Elementary teacher perceptions of school violence: Safe school elements and responsibility.

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content ELEMENTARY TEACHER PERCEPTIONS OF SCHOOL VIOLENCE:
SAFE SCHOOL ELEMENTS AND RESPONSIBILITY
by
Ivy Kazue Kelling
A Dissertation Presented to the
FACULTY OF THE ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF EDUCATION
August 2006
Copyright 2006 Ivy Kazue Kelling
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3236517
Copyright 2006 by
Kelling, Ivy Kazue
All rights reserved.
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
®
UMI
UMI Microform 3236517
Copyright 2006 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Acknowledgements
Ku ka lau lama i ke kukui ho ’ okahi.
This accomplishment is shared with all those who paved the path before me,
for those who stood beside me and with the continued vision of the future
generations to follow. Na ke Akua no i alaka’i mai ia’u a i ho’okele i keia wa’a.
Nana no ho’i i ho’olei ia’u me ke aloha o na kanaka i pili ai.
Ke komike loea. Mahalo to my dissertation committee, Dr. Melora Sundt,
Dr. Ron Astor and Dr. Stuart Gothold, for their time, commitment and profound
expertise. I am most honored to have worked with such a knowledgeable team and
appreciate the genuine aloha shared at each meeting.
Ko ’ u mau hoa kupa ’ a. I have had the pleasure of working and growing with
the two of the most motivated, persistent and hard-working mothers, students,
educators, leaders, and friends. Leigh Ann Siaosi and Karin Collier, your courage,
strength, brilliance, enthusiasm and passion will continue to inspire and motivate me.
It has been an unforgettable adventure and I am blessed to have been a part of it.
Mahalo to you both for getting on the same wa’a a pae akula.
Ka ‘ Ohana Kamakau. Mahalo nui loa ia ‘oukou pakahi a pau no ke kako’o
piha ‘ana mai ia’u a no ka hilina’i mau i ka waiwai o ka ‘imi ‘ana i ka na’auao. E na
keiki o Kamakau, mahalo ia ‘oukou pakahi no ke a’o ‘ana mai ia’u a no keia pilina
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
aloha. E ku’u mau tita e- Kalaunuola Domingo, Kehaulani Pe’a, Kaluapi’ilahaina
Cavaco a me ‘ Auli’i Costa, mahalo piha.
Ke Kumu Ho ’ ohalike. To Sheri Schonleber, my graduate instructor at
Chaminade University, who showed me the importance and role of “connectedness”
in school safety. You have been and continue to be an inspiration to my journey of
life-long learning and a model that I aspire to. I share this success with you. Mahalo.
Ku ’ u ‘ ohana aloha. To my ‘ohana, I am eternally grateful for your continued
support, your unconditional aloha, patience and for convincing me that the sky’s the
limit. The consistent words of encouragement have been greatly appreciated
throughout this process.
I ka ha o ke Akua, ua hanau ‘ia ka ‘ohana aloha la e. E ku’u kane aloha ‘o
Kaipo’i, mahalo ia ‘oe no ke kako’o mau ‘ana ia’u a no ka hilina’i i ka’u hiki. No
ka’u mau keiki, ‘o ‘A’ali’ihina’oleikamakani laua ‘o Kalehuakeahika’alani, no keia i
‘ike laua i ka waiwai o ka ‘imi na’auao mau ‘ana. Me ke aloha pau ‘ole no i hana ai.
Mahalo, mahalo, mahalo.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV
Table of Contents
Acknowledgements...........................................................................................................ii
List of Tables................................................................................................................... vi
Abstract........................................................................................................................... vii
CHAPTER 1 ......................................................................................................................1
Background of the Problem......................................................................................... 3
Purpose of the study....................................................................................................18
Research Questions and Hypotheses..........................................................................19
The Importance of the Study......................................................................................22
Limitations, Delimitations, Assumptions.................................................................. 24
Limitations................................................................................................. 24
Assumptions............................................................................................... 24
Definitions of Terms...................................................................................................25
CHAPTER 2 ................................................................................................................... 27
Protective and Risk Factors of Violence................................................................... 28
Figure 1: Elements of a Comprehensive Conceptual Model of Safe
Schools........................................................................................................32
Elements of a Safe School......................................................................................... 33
School Climate and Student Connectedness............................................ 33
High Expectations......................................................................................39
Collaborative Community Partnerships................................................... 42
Response Planning: Discipline and Crisis Plans......................................48
Responsibility for Providing Safe Schools...............................................................56
Conclusion.................................................................................................................. 61
CHAPTER 3 ................................................................................................................... 62
Introduction................................................................................................................. 62
Research Questions and Hypotheses......................................................................... 62
Research Design..........................................................................................................65
Demographics for Sample Schools...........................................................72
Instrumentation.......................................................................................... 74
Data Collection........................................................................................... 76
Data Analysis............................................................................................. 77
Limitations..................................................................................................77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4 ................................................................................................................... 78
Introduction................................................................................................................. 78
Sample..........................................................................................................................78
Research Questions, Hypotheses, and Results..........................................................79
Overall Perceptions of School Violence.................................................. 79
Existence of Safe School Elements...........................................................93
Responsibility for Safe Schools.............................................................. 103
CHAPTER 5 ..................................................................................................................108
Introduction................................................................................................................108
The Purpose of the Study..........................................................................................109
Summary of Findings................................................................................................109
Conclusions................................................................................................................113
Recommendations..................................................................................................... 117
References......................................................................................................................121
Appendix A: Teacher Questionnaire............................................................................131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
List of Tables
Table 1. Early Risk Factors and Effect Sizes for Predicting Violence at Age
15 to 18 and Protective Factors by Domain..........................................................31
Table 2. Demographic Data of Five Sample Schools.................................................. 73
Table 3. Teacher Response Rates by Gender, Years of Experience, Years at
School Site, and School Violence Level...............................................................79
Table 4. Percentage of Teachers' Perceptions of Observed Risk Behaviors in
School.......................................................................................................................81
Table 5 Teacher Perceptions of Violent Behaviors in School (percent).....................83
Table 6. Percentage of Teachers' Perceived Magnitude of School Violence by
Gender, Years of Experience, Years at Particular School Site, and Level
of School Violence..................................................................................................85
Table 7. Independent T-Test for Assaults, Thefts, and Verbal Abuse Scale Score... 87
Table 8 Independent T-test for Violent Behaviors Involving Weapons Scale
Score.........................................................................................................................87
Table 9. Comparison of Teacher Perceptions of Violent Behaviors in School
by Years of Teaching Experience......................................................................... 89
Table 10. Comparison of Teacher Perceptions of Violent Behaviors in School
by Years of Teaching Experience at Particular Site........................................... 91
Table 11. Comparison of Teacher Perceptions of Violent Behaviors in School
by School Violence Levels.....................................................................................92
Table 12. Comparison of Elementary School Teachers’ Perceptions of Safe
School Elements Across Violence Levels.............................................................97
Table 13. Percentage of Teachers’ Perceived Responsibility for Ensuring Safe
Schools...................................................................................................................104
Table 14. Percentage of Teachers’ Perceived Responsibility for Ensuring Safe
Schools by School Violence Level..................................................................... 106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Abstract
The purpose of this study was to examine the perceptions of elementary
school teachers about the extent of school violence in Hawai’i’s public elementary
schools. Teacher perceptions of violence were compared by gender, years of
teaching experience, number of years at the present school, and level of school
violence. This study also investigated the perceived existence of the five elements of
safe schools across schools with low, moderate and high violence levels as
determined by each school’s suspension rates. Finally, this study examined the
perceived responsibility of teachers for providing safe schools.
Data were collected with a structured survey consisting of 27 multi-part
Likert scale questions. The sample was comprised of 102 teachers from five
elementary schools. The results indicated that teacher perceptions varied
significantly by the number of years of teaching experience. Significant findings
were also found between the school violence levels for three of the safe school
elements: school climate, high expectations and discipline and crisis plans. Finally,
as the school violence level increased, teachers reported lower perceived
responsibility for teachers & parents while administrators were perceived with
primary responsibility.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
CHAPTER 1
INTRODUCTION
Headlines such as, “Shooting Rampage by Student Leaves 10 Dead on
Reservation” have recently emerged nationwide in response to the second most fatal
school shooting incident in U.S. history (Borja, 2005, March 22). A troubled 16-
year-old first shot and killed his grandfather and his companion before walking past
metal detectors into Red Lake High School, located on an isolated reservation in
Minnesota. He shot seven others before taking his own life.
Intense national media coverage of such tragic incidents and similar violent
acts on school campuses, however uncommon they may be, has raised public
concern over the past decade. Violent acts in schools, however, are surprisingly on a
downward trend (Devoe, Peter, Kaufman, Miller, Noonan, Snyder et al., 2004).
Despite the presence of a crisis management plan and drills conducted at
school, security guards and metal detectors, violence permeated through the Red
Lake Campus of 355 students. The questions that now arise are whether or not this
incident could have been prevented and who is liable. A review of statements made
by family members of the teenage shooter, administrators and peers of his school has
revealed that early warning signs were present (Borja, 2005, March 22; Borja &
Cavanagh, 2005, March 24; Johnson, 2005, March 25). The risk factors presaging
this shooting incident include a history of violence in his family and a lack of school
and peer connectedness as he was kicked out of school.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
This case indicates the need for more than the illusion of physical protection
offered by security guards and metal detectors. Rather, a need exists for a more
comprehensive prevention plan that addresses other significant factors such as
school, peer and family connectedness and a positive school climate where students
and teachers are trusted and respected (King, Vidourek, Davis, & McClellan, 2002;
Smith & Sandhu, 2004; Web & Norton, 2003; Wilson 2004).
Additionally, media coverage of recent brawls at several Hawai’i public
school campuses in the spring of 2005 has increased public and community concern
and amplified the fear of students (Fujimori & Martin, 2005). This increased
apprehension of parents and communities revealed the reporting discrepancies of
violence in Hawai’i public schools (Brannon, 2005). The magnitude of violence on
Hawai’i school campuses was difficult to assess after a change in the state tracking
system of violent incidents resulted in lost data (Brannon, 2005). Moreover, a search
of school violence studies conducted in Hawai’i resulted in limited resources. Such
studies by Chesney-Lind and Nakano (2003) and Grunbaum et al. (2002) focused on
middle and high school student self-reports, overlooking the prevalence of violence
in Hawai’i’s elementary schools as perceived by teachers.
This study is one of three studies that will assess violence and perceived
safety in all three levels of education in the most populated island in the state of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
Hawai’i. This study singles out elementary schools to ask about teacher perceptions
of violent behaviors in the schools and the presence of five safe school elements that
comprise the comprehensive prevention model in this study.
Background of the Problem
History reveals that violence pervades American society with explicit
examples of civil wars and the suppression of minority races. Schools are not
immune to violence and serve, in many cases, as the setting for aggressive acts
between students. Despite the media stories, schools are showing a positive promise
to tackle violence by confronting the problem with prevention and intervention
strategies as well as through team partnerships between those involved as
stakeholders.
With intense media coverage focused on the extreme cases of violence in
American schools, the public is being misled by the notion that school violence is
worsening. Apprehension by the concerned public in addition to legislation based on
changes to federal policies continue to drive numerous national studies on the topic
of school violence. Most recent studies conducted by government agencies such as
the Bureau of Justice Statistics (2004) and the National Center for Education
Statistics (Miller, 2000) show, using self-reported perceptions of school
administrators, teachers, students and parents, a general trend of American schools
becoming safer learning environments. The concerned community, however,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
continues to motivate the accountability movement towards providing safe learning
environments as the literature expands to include more types of comprehensive and
multidisciplinary approaches. The following section lays the foundation of violence
in schools, beginning with a definition of school violence and is followed by the
most recent national and Hawai’i trends, incidence and victimization rates.
School Violence Definition.
The definition of “school violence” is broad and dependent on the perspective
of each particular study or author working with the issue. Furlong and Morrison
(2000) contend that the term school violence has evolved over the past decade since
the early 1990s. They state, “School violence is now conceptualized as a
multifaceted construct that involves both criminal acts and aggression in schools,
which inhibit development and learning, as well as harm the school’s climate”
(P-71).
In the process of defining school violence, Furlong and Morrison (2000)
stress the importance of differentiating between “school violence” and “violence in
the schools” (p. 73). The violence in the former label occurs when school is the
context where problems are caused whereas the latter example occurs when
problems escalate in school as a result of situations outside of school. Although the
causes may be different, the setting remains the same and needs to be nonetheless
addressed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
The Violence in U.S. Public Schools: 2000 School Survey on Crime and
Safety (Miller, 2003) defines violence as, . .actual, attempted, or threatened fight or
assault” (p. 100). This study, however, failed to include other crimes such as theft
and sexual assaults as are further covered by the Indicators of School Crime and
Safety: 2004 (2004). The School Survey on Crime and Safety defines violent
incidents as, “.. .rape, sexual battery other than rape, physical attacks or fights with
or without a weapon, threats of physical attack with or without a weapon, robbery
with or without a weapon” (Miller, 2003, p. 168).
Benbenishty and Astor (2005) support Furlong and Morrison’s (2000) earlier
position with a comprehensive theory that includes all stakeholders dealing with
victimization and the effects on their emotional, cognitive, and social domains.
Benbenishty and Astor (2005) used the following broad definition of school violence
and victimization,
.. .any behavior intended to harm, physically or emotionally, persons in
school and their property (as well as school property). We define
victimization as a student’s report that another student or staff member
perpetrated school violence against him or her. This broad definition includes
verbal and social violence (such as cures, humiliation, social exclusion);
threatening behaviors (direct, indirect, extortion, scary behavior); physical
violence (such as pushes, kicks, punches, beating); stealing and damaging
property; weapon use (carrying, threatening, using); and sexual harassment.
(P- 8-9)
This, thus far, has been the most inclusive definition of violence and will be used for
the purpose of this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
Violence Incidence Rates in American Schools
As noted earlier, violent incidents in America’s schools are on a downward
trend (Devoe et al., 2004). The most current government statistical study on school
violence is the Indicators o f School Crime and Safety: 2004 (Devoe et al., 2004).
This comprehensive annual report is a compilation of several national surveys of
school principals, teachers, and students nationwide. Data are arranged according to
a total of 19 indicators of school crime and safety (Devoe et al., 2004). The findings
of this report were disaggregated by student fatal and non-fatal victimization,
violence and crime at school, nonfatal teacher victimization at school and school
environment.
Violent deaths at school in the year 2001-2002 accounted for 17 school-aged
children from ages 5-19, compared to 32 deaths in 1999-2000 (Devoe et al., 2004).
Fourteen of the deaths in 2001-2002 were attributed to homicides and the remaining
three to suicides. This is a 42 percent decrease from reports in 1997-98 when 40
school-aged children were killed in a school related death (Devoe et al., 2004).
Although these numbers may continue to alarm concerned parents, national data on
school-aged children reflect schools as safe havens. Devoe et al. (2004) states, “In
each school year from July 1, 1992, to June 30, 2000, youth ages 5-19 were at least
70 times more likely to be murdered away from school than at school” (p. 6). This
accounts for 24,406 children ages 5-19 who were victims of homicide and 16,735
children who committed suicide outside of school between 1992 and 2000.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
Nonfatal student victimization also improved and rates show a decreasing
trend over the past decade. From 1992 to 2002, the violent victimization rate at
school decreased by 50 percent, “from 48 to 24 crimes per 1,000 students at school
and from 71 to 26 crimes per 1,000 away from school” (Devoe et al., 2004, p. 10).
The findings also show the vulnerability of younger students (ages 12-14) as they are
more likely victims at school compared to their older counterparts (ages 15-18)
(Devoe et al., 2004, p. 10). It is no surprise then that the percentage of nonfatal
crimes of students ages 12-18 reports a decrease by 50 percent from 1995 to 2003
(from 10 percent to 5 percent) (Devoe et al., 2004). A slight decline in the percentage
of 9th -12th grade students is shown between 1993 and 2003 (16% and 13%
respectively) as males participated in an alarming 47 percent higher level of violent
crimes than females; discriminating existing gender differences.
One indicator in the Indicators o f School Crime and Safety: 2004 report
(Devoe et al., 2004) shows an increase in bullying incidence rates at school. Bullying
is defined to include the following acts, “picked on or made to do things they did not
want to do” (Devoe et al., 2004, p. 20). Students ages 12-18 responded in the School
Crime Supplement to the National Crime Victimization Survey. Victims of bullying
increased from 5 percent in 1999 to 8 percent in 2001 and slightly decreased to 7
percent in 2003. Age and grade level were inversely related to student victimization
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
of bullying in schools. Reports in 2003 reveal that “14 percent of 6th -graders,
7 percent of 9th -graders, and 2 percent of 12th -graders reported that they had been
bullied at school” (Devoe et al., 2004, p. 20).
The Wide Scope Questionable Quality study conducted by the U.S.
Department of Education is a summary of three reports that examined the topic of
school violence and prevention in American schools (Crosse, Burr, Cantor, Hagen, &
Hantman, 2002). Surveys assessed the perceptions of 886 K-12 school principals in
1997 and 1998 and the perceptions of secondary teachers and students in 1998. The
results are relatively high levels of violent crime and fewer levels of serious violent
crimes (Crosse et al., 2002). Principal reports show, “...66 percent of the schools
experienced one or more incidents of less serious violent crime or property crime
(i.e., fighting without a weapon, vandalism, or theft) and 10 percent experienced at
least one serious violent crime (i.e., fighting with a weapon or robbery)” (p. 2).
The data also report more violent incidents occurring in middle schools than
either elementary or secondary schools. The greatest difference between school
levels is evident in the incidents of physical attacks or fights with a weapon. Twenty-
one percent of middle school principals reported physical fights with a weapon
compared to 2 percent in elementary and 11 percent in high schools (Crosse et al.,
2002). In comparison, fights without a weapon occurred at 72 percent of middle
schools, 56 percent of high schools and at 34 percent of elementary schools. In both
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
cases, middle schools were perceived by principals to have the highest levels of
problem behavior, followed by high schools and then elementary schools.
The results of the Wide Scope Questionable Quality report should, however
be reported with caution. First, the definition used in this study for “serious crime”
includes crimes reported to police or law enforcement officials, one much more
restrictive than what is reported by others. Second, principals are often not aware of
every incident that occurs on campus except the most severe and thus, perceptions
may differ from those of teachers and students. Teacher and student perceptions were
included in the following year in 1998 for a more comprehensive study that focused
on student and teacher victimization in the middle and high schools.
A second study based on principal perceptions and conducted by the National
Center for Education Statistics (Miller, 2000) reports quite similar findings through
the perceptions of 2,270 school principals from regular public K-12 school sites
nationwide. The major findings indicate that 71 percent of public schools
experienced at least one violent incident during the 1999-2000 school year and 20
percent experienced at least one or more serious violent crime. Definitions used in
this study for the terms of violent incident and serious violent incident are identical
to those used in the Crosse et al. study (2002).
Miller (2000) reports positive correlations of violence to the following
factors, secondary school (more likely than elementary schools to report incidents),
size of school’s enrollment, urbanicity, high value placed on education, number of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
classroom changes, number of serious disciplinary problems, and high amounts of
class disruptions (anthrax, bomb threats). Results from the 2000 School Survey on
Crime Supplement differ from the survey two years earlier in the prevalence of
violent incidents reported between the school levels. Secondary schools reported the
most violent incident (92%) than either middle (87%) and elementary (61%) schools
and an equal percentage of serious violent incidents with middle schools (29%)
(Miller, 2002). However, once again this study is based on the perceptions of the
principal and does not include those of the teachers or students and should be taken
with caution.
Much of the violent incidence rates from national studies in American
schools are a result of self reports and reveal the perspective of both school
principals and students and most recently of teachers. Up until a decade ago, school
violence research was conducted by disciplines other than education and results
reveal a disconnect between what is occurring on campuses and incidents that are
being reported by the school administration (Furlong &Morrison, 2000). It is,
therefore, critical to examine the perceptions of teachers about violence in the
schools as it may affect their performance in the classroom and their relationships
with the students (Zeira, Astor, & Benbenishty, 2004). In addition, teacher
perceptions are essential for determining effective prevention and intervention
strategies to reduce violence in schools.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
Although self-reports may reveal one’s thoughts and feelings, they should
also be taken with caution. Self reports may not always represent reality. Recent
reliability and validity checks of such self reports reveal that those who fail such
checks are providing significantly higher rates of violent victimization than the
comparison group (Furlong & Morrison, 2000). Rosenblatt and Furlong (1996)
conducted a reliability check and the results indicate that participants who failed the
test, “reported more victimization,.. .higher perceptions of dangerous conditions on
the school campus,.. .lower grades,.. .and fewer peer connections” (p. 193).
Therefore, statistical information based on self-reports must be interpreted with
caution.
A summary and overview of the most recent national reports reveal that
despite improvements, violence in schools is still prevalent and is continuing to drive
further research. However, national trends alone do not affirm incidence of violence
for all states or neighborhoods. Rather, the incidence of violence must be considered
at the individual state level. The next section presents an overview of violent
incidence rates in the state of Hawai’i in comparison to national trends.
Violence Incidence Rates in Hccwai ’ i
Although there has been no fatal school shooting incident in Hawai’i, it has
not been without violent incidents on school campuses. Local Hawai’i newspaper
headlines recently read, “Campus fighting strikes 3 schools,” and “Threat of violence
paralyzes Radford” (Leone, 2005, February, 5; Martin, 2005, February 1). With such
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
media coverage, the surrounding communities’ apprehension has not been eased and
may have been compounded with headlines that read “School violence difficult to
assess” (Brannon, 2005, February 15). The literature reports an ambiguity of current
statistics reported for Hawai’i’s school violence in comparison to previous years.
The following statement was made in a local newspaper, “The system for reporting
student discipline was changed a few years ago, and some data went unrecorded or
were simply lost during the transition, resulting in artificially low statistics for some
previous years” (Brannon, 2005, February, 15).
The Superintendent’ s Fourteenth Annual report on School Performance and
Improvement in Hawaii 2003 (Department of Education State of Hawai’i, 2004)
verifies that in 2001-2002, unfamiliarity of a new computer reporting system resulted
in lost data and about 3000 fewer reported incidents than the previous year.
Additionally, the ambiguity of federal reporting requirements creates a difficult
situation for portraying the most accurate image of school safety in Hawai’i
(Brannon, 2005; Essoyan, 2003; Kingery & Coggershall, 2001).
The 2003 Superintendent’s Report (Department of Education State of
Hawai’i, 2004) also shows the reliance on suspension charges to reveal current safety
in the schools, which overlooks the less severe non-suspendable behaviors. Hawai’i
suspension rates are divided into four categories and include violence (assault,
dangerous weapons, extortion, firearms, murder, robbery, sexual offenses, terroristic
threatening and harassment), property (burglary, property damage, theft, and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
trespassing), illicit substances (alcohol use or possession, drug paraphernalia,
marijuana use or possession, other illicit substance use or possession, sale of illicit
substances, smoking or tobacco contraband), and order (disorderly conduct, false
alarm, gambling, insubordination, laser pen or pointer, other prohibited conduct). All
categories showed the highest incidence of offenses in 1995-1996 and have steadily
declined until 2002-2003 with the exception of violence. The slight increase of
violent incidents in 2002-2003, however, may be contributed to the lost data in 2001-
2002 and possibly to a variation in the implementation of suspensions.
Discrepancies in state reporting are evident across reports and studies. One
such area of difference is that of weapon and gun possession. The most current
Hawai’i report to the U.S. Department of Education suggests the highest incidence of
guns found in state schools was during the 2003-2004 school year (Brannon, 2005;
Department of Hawai’i, 2005). Hawai’i reported a total of six firearm incidents
statewide during this year. This rather small number misleads the public from the
total of 47 cases that also include 35 air guns, two explosives and one other type of
firearm (Department of Education State of Hawai’i, 2005).
Yet in comparison to the national prevalence of gun incidents on campus,
Hawai’i appears to be safe. The 1999 School Crime Supplement to the National
Victimization Crime Survey reported that in 1999, 7.5 percent of students nationwide
between the ages of 12 and 18 knew a student who brought a gun to school, while
2.8 percent saw a student with a gun at school (Addington, Ruddy, & Chandler,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
2002). These numbers show a drop from 1995 when 12.7 percent of students
reported knowing a student who brought a gun to school and 5.3 percent saw a
student with a gun at school (Addington et al., 2002).
A second national self report survey from the United States Youth Risk
Behavior Survey of 2001 reports that 17.4 percent of high school students carried a
weapon (gun, knife or club) and 5.7 percent of high school students reported carrying
a gun (Grunbaum et al., 2002). In the same study, Hawai’i, in comparison reported
that 10.6 percent of high school students carried a weapon and 2.9 percent of high
school students carried a gun to school (Grunbaum et al., 2002). For both weapon
and gun carrying rates, Hawai’i fares well below the national average in this
particular study. This, however, does not leave Hawai’i victimless to less severe
violence in schools (i.e, physical assault and theft).
The United States Youth Risk Behavior Survey o f2001 reports that in 2001,
6.6 percent of the nation’s high school students felt too unsafe to attend school and
12.5 percent engaged in a physical fight on school campus compared to 37 percent in
1997 (Grunbaum et al., 2002). Similar results are given for Hawai’i high school
students who are too afraid to attend school (6.7%), while 9.0 percent of Hawai’i
high school students were involved in a physical fight on campus, a statistic that
dropped from 11.5 percent in 1999 (Essoyan, 2003; Grunbaum et al., 2002). This
data, however, is inconsistent with the figures presented in the Hawai ’ i Department
o f Education Trend Report (2005).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
The Trend Report (Hawai’i Department of Education, 2005) presents a total
of 32 violent incidents per every 1,000 high school students during the 2003-2004
year. In addition, four violent incidents per every 1,000 elementary students are also
reported. Nonetheless, these statistics reveal that although Hawai’i schools appear
safer than their mainland counterparts, students in Hawai’i deem school campus
equally unsafe to attend. Crime in Hawai’i however, illustrates a frightening picture
of youth violence in the state (Gao & Perrone, 2004). Alarming rankings compared
to other states is reason for concern for school-aged children as such behavior may
transfer into the schools.
Gao and Perrone (2004) report that Hawai’i’s index crime rate alarmingly
ranks second in the nation. This is attributed to having the highest larceny rate in the
nation and a high overall property crime rate. Violent crimes in Hawai’i as defined to
include negligent manslaughter, other assault and sex offenses accounted for 1,184
juvenile arrests in 2003 and comprise 36 percent of all arrests made for violent acts
(Gao & Perrone, 2004). A majority of the violent crimes were committed by
juveniles between the ages of 13-14 (Gao & Perrone, 2004). These ratings suggest a
drastic need to address the issues of violence and to curb numbers of further growth.
In addition to high crime index rates and underreporting of violence against
teachers (Essoyan, 2003, online), violence in Hawai’i’s schools may be elevated by
overpopulation. A body of research support smaller school enrollment in addressing
violence in the schools (Chesney-Lind, 2003; Larsen, 2003; Miller, 2000). Hawai’i’s
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
public schools are on average larger than the national average (Chesney-Lind, 2003;
Hawai’i Educational policy Center, 2003). The Hawai’i Educational Policy Center
(2003) notes that Hawai’i’s elementary schools are 23.6 percent larger than the
national average, middle schools are 40.5 percent larger and high schools are 95.2
percent larger than their national counterparts (Chesney-Lind, 2003; Hawai’i
Educational Policy Center, 2003). Overpopulated public schools in Hawai’i must be
considered in tackling school violence.
Hawai’i related data on school violence in the elementary schools are
minimal, while numerous studies have focused on middle and high school personnel
and students. The primary grades have been neglected in the research. The following
section will briefly review the trends of violence in elementary schools as revealed
by the limited research.
Violent Incidence in Elementary Education
Elementary schools have in most cases shown the fewest incidents of violent
acts and crimes compared to their counterparts in education. Research shows an
inverse relationship between school level and violent crimes involving both student
and teacher victimization (Devoe et al., 2004; Crosse et al., 2000). One exception is
that elementary schools present higher occurrences of bullying than found in either
middle or high schools (Addington et al., 2002; Devoe et al., 2004; Furlong, &
Morrison, 2000).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
The 1999 School Crime Supplement Report indicates that 10.5 percent of
6th graders perceived themselves as victims of bullying compared to 1.2 percent of
12th graders (Addington et al., 2002). Batsche and Knoff (1994) state that regardless
of age, the youngest on campus is often the innocent target of bullies. This is logical
considering the definition of bullying as “.. .a form of aggression in which one or
more students physically and/or psychologically (and more recently, sexually) harass
another student repeatedly over a period of time” (Batsche & Knoff, 1994, p. 165).
These studies, however, include only students in grades 6-12 and should therefore be
taken with caution.
Astor (1995) acknowledges that most violence intervention programs are
focused on adolescents while the younger children are not considered. Research has
shown, however, that aggressive young children are predictive of aggressive
behavior in adolescence and adulthood (Astor, 1995). Olweus’ 22 year longitudinal
study (1991) found that 60 percent of boys who were peer-nominated as bullies in
grades 6-9 had at least one conviction by age 24. Even more alarmingly, up to 40
percent of those nominated as bullies had three or more convictions by the same time
in comparison to 10 percent of the non-aggressive boys. Both Astor and Olweus
concur that the fate of aggressive children can be changed with early intervention,
starting in the elementary school years.
The research also establishes that school violence is positively related to the
following factors: size of school enrollment, location of the school (urban vs. rural),
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
number of classroom disruptions, and number of classroom changes (Larsen, 2003;
Miller, 2000). This would suggest that the safest elementary schools are those with a
smaller enrollment size, located in a rural area, with minimum classroom disruptions
and classroom changes. These characteristics would result in a less significant
number of violent incidents on campus than a school of the reverse situation (Larsen,
2003; Miller, 2000;).
Violence pervades school campuses across all levels and grades. It is
therefore important to discriminate the significant differences between them to
appropriately act and address specific problems at each of the various school levels.
It is obvious that a one-size-fits-all approach to school violence would not succeed if
implemented at all levels in equal or similar ways. Rather, an approach that is more
focused on the specific needs of each level to address particular crisis areas is a more
effective method.
Purpose of the study
The purpose of this study was to examine the extent of school violence in
Hawai’i’s public elementary schools through self-reports of teachers. School
personnel perceptions of violence were compared by gender, years of teaching
experience, number of years at the present school, and level of violence present at
the school. This study also examined the extent to which teachers perceive the
existence of the five elements of a comprehensive conceptual model of safe schools—
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
elements that have repeatedly emerged throughout the literature and in successful
prevention programs. These elements include the following: a positive school
climate, connectedness, high expectations, collaborative community partnerships and
discipline and crisis plans.
Finally, this study examined the perceived responsibility of teachers for
providing a safe school. As recent media coverage of fatal school shootings has
raised a concern for accountability measures within the schools, the question then
arises, “Who is responsible for providing a safe school?”
Research Questions and Hypotheses
Perceptions o f School Violence
Research Question 1: What are teachers’ perceptions of observed risk
behaviors, violent behaviors in school, and overall magnitude of school violence in
elementary schools on O’ahu? Are there differences in perceptions of observed risk
behaviors, violent behaviors in school, and overall magnitude of school violence by
gender, years of teaching experience, years of teaching at school site, or violence
level?
Research Question la: What are teachers’ perceptions of violent behaviors in
school, and overall magnitude of school violence in O’ahu elementary schools?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
Research Question lb. Are there differences in the perceptions of violent
behaviors at school by gender, years of experience, years at particular school site or
level of violence?
Hypothesis lb: Females will report a greater frequency of violent behaviors
at school than males. Less experienced teachers will report greater frequency of
violent behaviors than veteran teachers. Less experienced teachers at a particular
school site will report a greater frequency of violent behaviors than veteran teachers
at that same site. With increased school violence levels, greater occurrences of
violent behaviors will be reported.
Perceived Existence o f Safe School Elements
Research Question 2: Is there a significant difference in elementary school
teachers’ perceptions of the five safe school elements (school climate,
connectedness, expectations, collaborative community network, and discipline/crisis
plans) between teachers from low violence, moderate violence, and high violence
schools on O’ahu?
Research Question 2a: Is there a difference in school climate scores for
teachers from low, moderate, and high violence schools?
Hypothesis 2a: As the school violence level increases, less positive school
climate attributes will be reported along with more negative school climate
attributes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
Research question 2b: Is there a difference in connectedness scores for
elementary school teachers from low, moderate, and high violence schools?
Hypothesis 2b: As the school violence level increases, teacher
connectedness scale scores will decrease.
Research question 2c: Is there a difference in expectations scores for
elementary school teachers from low, moderate, and high violence schools?
Hypothesis 2c: As the school violence level increases, teacher expectations
scale scores will decrease.
Research question 2d: Is there a difference in collaborative community
networking scores for elementary school teachers from low, moderate, and high
violence schools?
Hypothesis 2d: As the school violence level increases, teacher collaborative
community networking scale scores will decrease.
Research question 2e: Is there a difference in discipline and crisis plan scores
for elementary school teachers from low, moderate, and high violence schools?
Hypothesis 2e: As the school violence level increases, teacher discipline and
crisis plans scale scores will decrease.
Perceived Responsibility for Safe Schools:
Research question 3: To what degree do elementary school teachers perceive
the stakeholders (administrator, teachers, parents, students, community, resource
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
teachers and complex area supervisors, and DOE state level supervisors/personnel)
responsible for contributing to safe schools?
Research question 3a: How does perceived responsibility for providing safe
schools vary amongst schools with low, moderate, and high reported violent
incidents?
Hypothesis 3a: As the school violence level increases, teachers report lower
perceived responsibility for each of the stakeholders (administrator, teachers,
parents, students, community, resource teachers and complex area supervisors, and
DOE state level supervisors/personnel).
The Significance of the Problem
Although numerous national studies have focused on teacher and student
perceptions of violence in schools nationwide, Hawai’i-specific data are limited and
most do not include the perceptions of frontline workers. Hawai’i-related data rely
on results from the Hawai’i Youth Risk Behavior Survey that is distributed to all
regular middle and high school students in Hawai’i during the spring semester of
each odd-numbered years (Saka & Bunao, 2005). Hawai’i also relies on overall state
suspension rates to reveal the safety condition of its schools.
The review of the literature produced a limited amount of research on
elementary school personnel perceptions of violence, however, none was specific to
Hawai’i. The need exists for a focus on elementary school teachers to gain a more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
complete understanding of the pervasiveness of violence in Hawai’i public
education. In addition, although research has found risk factors of violence in the
schools, there has not been a study that investigated the presences of these elements
as proposed in this study’s comprehensive conceptual model.
This study examined the teacher perceptions of school violence in Hawai’i
and how they vary by gender, total years of teaching experience, length of
employment at a particular site, and level of school violence. It also addressed the
perceptions of the existence of five safe school elements to include school climate,
connectedness, high expectations, collaborative community networking, and
discipline and crisis plans. This study will add to the current research and further
indicate the importance of school and teacher characteristics in affecting school
safety through effective prevention and intervention strategies.
Lastly, this study examined teacher perceived responsibilities for assuring
safe campuses. Results will better inform teacher, parent and community training
programs about expectations for assumed responsibility while revealing the beliefs
behind current action in education. Results will also inform policymakers and
administrators of the areas of need for future training and improvement of existing
prevention and intervention strategies.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Limitations, Delimitations, Assumptions
Limitations
1. Data were limited to teacher perceptions and exclude the perceptions of
students and administrators
2. The level of school violence in this study (low, moderate and high) was
determined by the percentage of student population suspended during the
2003-2004 school year.
3. Teacher perceptions were limited to the five schools randomly selected to
represent each of three levels of school violence.
4. The study was limited to the response rate and to the sample who
volunteered their participation.
Assumptions
1. Participating sample responded honestly to the study instrumentation.
2. The level of school violence in this study (low, moderate and high) can
reasonably be determined by the percentage of student population
suspended.
3. Previous studies completed reliability and validity checks with the
questionnaire questions adapted for this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
Definitions of Terms
School Violence was previously defined to include Benbenshity and Astor’s
(2005) definition,
.. .any behavior intended to harm, physically or emotionally, persons in
school and their property (as well as school property). We define
victimization as a student’s report that another student or staff member
perpetrated school violence against him or her. This broad definition includes
verbal and social violence (such as cures, humiliation, social exclusion);
threatening behaviors (direct, indirect, extortion, scary behavior); physical
violence (such as pushes, kicks, punches, beating); stealing and damaging
property; weapon use (carrying, threatening, using); and sexual harassment,
(p. 8-9)
In addition, each of the five safe school elements will be defined in chapter 2.
They include, school climate, connectedness, high expectations, collaborative
community networking and discipline and crisis plans.
Organization of the Study
Chapter 1 has presented the introduction, the background of the problem, the
statement of the problem, the purpose of the study, the research questions to be
answered, the hypotheses, the significance of the study, the limitations and the
definition of terms.
Chapter 2 reviews the literature and includes the following topics:
1. Protective and risk factors of violence
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
2. Safe school elements and a comprehensive safe school model (a positive school
climate, connectedness, high expectations, a collaborative community
partnership, and discipline and crisis plans)
3. Responsibility for providing safe schools
Chapter 3 presents the methodology used in this study. It includes the
research design, population and sampling procedures, instrumentation development,
data collection and a section on validity and reliability. The chapter concludes with
methods of data analysis and the limitations.
Chapter 4 presents the results of the study through the use of descriptive and
inferential statistics.
Lastly, chapter 5 concludes with a summary of the findings and suggestions
for applied practice and for future research.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
CHAPTER 2
LITERATURE REVIEW
Despite reports that show a decline of violence in America’s schools, . .U.S.
society has the greatest increase in the number of programs supported by research”
(Astor, Meyer, Benbenshity, Marachi & Rosemond, 2005, p. 20). Yet, Hawai’i
specific data remains limited as violence in local schools persist. In addition, the
topic of school violence in Hawai’i has focused on the adolescent ages while
neglecting the elementary years. To understand the magnitude of the school violence
problem in the early years it is essential to recognize the predictors of violence at
these younger ages as well as the key protective factors that reduce violence.
This chapter reviews the literature of school violence and the studies that
contribute to creating safer schools through the implementation of a comprehensive
safe school plan. The first section examines the root causes of juvenile violence with
a review of protective and risk factors. A review of the literature about a
comprehensive conceptual model for building safe school settings comprises the
second section. This section includes the five safe school elements and principles for
their inclusion in the comprehensive conceptual model. Finally, the last section
explores the idea of responsibility and accountability for providing safe schools.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
Protective and Risk Factors of Violence
A substantial body of literature addresses risk and protective factors of school
violence. Protective factors are those that deter the likelihood of violence or
disruptive behavior while risk factors are defined to include variables that promote
and induce the likelihood of violence or disruptive behaviors (Schwartz, 1999;
Sprague, Colvin, Irwin, & Stieber, 1998).
A result of protective factors is increased student resiliency (Schwartz, 1999).
A sample of K-12 public school principals in Oregon reported the top rated
protective factors in their schools to include teacher-student relationships, school
climate, discipline, high expectations and supervision (Sprague et al., 1998).
Additional suggestions of protective factors include “...components to increase
understanding and acceptance of diversity, improve school climate and safety,
improve interpersonal relationships among students and between students and
educators, promote clear and consistent rules, and promote parent and community
involvement” (Hunt, Meyers, Davies, Meyers, Grogg, & Neel, 2002, p. 400).
In the aftermath of several fatal school shooting incidents, President Clinton
requested the development of early warnings against violent episodes. The United
States Department of Education, in response, developed the Early Warning Timely
Response, a compilation of research based practices to recognize and help those
students in need (Dwyer et al., 1998). Dwyer et al. (1998) state the necessity of
school connectedness with students, one of the protective factors, as a means to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
recognize the following 16 risk factors; social withdrawal, excessive feelings of
isolation, feelings of rejection, being a victim of violence, feelings of humiliation,
poor academic interest and achievement, demonstrated violence in writings and art,
uncontrolled anger, aggressive behavior, reoccurring discipline problems, a history
of violence, prejudicial and racial attitudes, substance abuse, gang membership,
access and use of firearm and serious threats of violence (p. 17). School staff,
families, and the community must responsibly make personal connections with each
student to recognize their individual needs and feelings.
Other risk factors may be specific to particular situations. The results of the
survey by Sprague et al. (1998) reported the top risk factors identified as present in
school campuses in Oregon are poverty, transiency, child abuse, truancy and
bullying. Astor, Benbenishty, Pitner and Meyer (2003) present risk factors of
bullying and victimization. These risk factors of school characteristics and bullying
victimization include location and time, school climate and awareness (Astor et al.,
2003). Astor et al. reported that bullying have been found to occur at specific times
and locations on campus and add that negative school climate and the awareness of
teachers along with their reluctance to help are risk factors for bullying in the
schools.
Caution is needed when considering risk factors as early warning signs of
violence. Warnings are given against false stereotyping of those who possess such
risk factors and assumptions that all children of a particular race, economic status or
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
appearance are susceptible to violence and aggression (Dwyer et al., 1998). Rather,
the use of wise judgment is encouraged in making decisions that will ultimately
affect the lives of children (Schwartz, 1999).
Other studies have separated risk and protective factors into five domains
including individual, family, school, peer-related, community/neighborhood, and
situational factors (Bennett-Johnson, 2004; Hawkins, Herrenkohl, Farrington,
Brewer, Catalano et al., 2000; Resnick, Bearman, Blum, Bauman, Harris et al., 1997)
Predictors of violence resulted from an analysis of the Seattle Social Development
Project longitudinal data set (Hawkins et al., 2000). Lipsey and Derzon (as cited in
Hawkins et al., 2000) studied predictors of serious adolescent delinquency and early
adulthood. They collected previous published and unpublished data and analyzed the
data to find relationships between the predictor variables and the criterion variable.
Lispey and Derzon also examined the relationships of predictor variables at
different age groups including ages 6-11 and ages 12-14. For the purposes of this
study of elementary teacher perceptions of violence, Table 1 includes early risk and
protective factors during the ages of 6-11 for violence onset at age 15-18. Included
are also the estimated aggregated effect sizes as an indication of significance
(Hawkins et al., 2000). Risk and protective factors are divided into five domains
including the individual, family, school, and peer group.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
Table 1
Early Risk Factors and Effect Sizes for Predicting Violence at Age 15 to 18 and
Protective Factors by Domain
Domain
Early Onset
(ages 6-11) Protective Factors
Individual General Offenses (.38)
Substance use (.30)
Being male (.26)
Aggression (.21)
Psychological condition
hyperactivity (. 15)
Problem (antisocial)
behavior (.13)
Exposure to television violence
Medical, physical (.13)
Low IQ (.12)
Antisocial attitudes,
beliefs
Dishonesty in males
Intolerant attitude toward deviance
High IQ
Being female
Positive social
orientation
Perceived sanctions
for transgressions
Family Low socioeconomic
status/poverty (.24)
Antisocial parents (.23)
Poor parent-child
relations- harsh, lax, or
inconsistent discipline (.15)
Broken home (.09)
Separation from
parents
Abusive parents (.07)
Neglect
Warm, supportive
relationships with
parents or other
adults
Parents' positive
evaluation of peers
Parental monitoring
School
Poor
attitude/performance (.13)
Commitment to school
Recognition for
involvement in
conventional
activities
Peer Group
Weak social ties (.15)
Antisocial peers (.04)
Friends who engage in
conventional behavior
Adapted from Potter, Beaulieu-Cooke, Davis, Hasbrouck, Rosenfeld,
Tuma, (2001) and Hawkins et al. (2000)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
Table 1 suggests practical implications for prevention planning and efforts.
The probability of violence in later stages of life may be reduced as a result of
increased appropriate protective factors in children ages 6-11 (Potter et al., 2001). In
addition, a preventative approach would address these predictive risk behaviors prior
to the developmental stages that they typically occur.
Figure 1: Elements of a Comprehensive Conceptual Model of Safe Schools
Intervention:
Discipline
and Crisis
Plans
Prevention:
Positive School Climate
Connectedness
High Expectations
Collaborative Community
Networking
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
Elements of a Safe School
From the previous review of risk and protective factors of school violence, a
comprehensive conceptual model for safe schools was developed. Throughout the
literature on comprehensive school plans addressing the incidence of violence on
campus, five themes emerged. These five essential elements include a positive
school climate, connectedness, high expectations, collaborative community
partnerships, and discipline and crisis plans. They are each further identified and
discussed in the following section. Figure 1 illustrates these elements and their
application to a systematic approach to school violence.
School Climate and Student Connectedness
Within suggestions made for a more integrated systemic approach to school
safety, recommendations are provided for three main components; prevention,
intervention and response planning (Dwyer & Osher, 2000; Pollack & Sundermann,
2001). Included in the prevention component of this framework lies a vast body of
literature discussing the relationship between positive school climates and school
safety, however, most are focused on the middle and high school populations and
exclude elementary school participants. Much of the existing elementary related
literature on school climate and connectedness are associated to bullying and to the
research on protective factors which contend that early intervention in the younger
grades helps to deter violence in later adolescent years (Astor, 1995; Hawkins et al.,
2000; Potter et al., 2001). Therefore, in addition to the literature aimed at elementary
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
schools, several studies of middle and high schools will also be reviewed in this
section to demonstrate possible results of ignoring violence and aggressive behaviors
in the elementary schools.
Webb and Norton (2003) define school climate as . .the collective
personality of a school or school system. It is the atmosphere that prevails as
characterized by the social and professional interactions of people” (p. 106). It
includes the attitudes and values of the school staff, students, parents and community
members (Haynes, 1996). Common characteristics of a positive school climate as
reported in the research include effective communication and positive relationships
between students and teachers, mutual respect and trust for school staff and peers, a
focus on academic achievement, discipline policies that are both fair and consistent,
attention to safety issues, and the involvement of family and the community (Webb
& Norton, 2003; Wilson 2004). A positive school climate additionally includes
positive team building as a means to work towards set goals and often encompasses
the element of school connectedness, defined by a student’s enjoyment of school and
a feeling of belonging as a result of apparent caring from teachers and peers (Fein,
Vossekuil, Pollack, Borum, Modzeleski, & Reddy, M., 2002; Resnick, Harris, &
Blum, 1993; Wilson, 2004).
The relationship between school climate and school connectedness is
common throughout the literature. Dwyer & Osher (2000) describe a critical
movement of creating caring school communities where members interact in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
supportive relationships, as one of four essential components of an effective school
plan. This concept demonstrates the significance of interpersonal relationships in
school environments and is replicated throughout the literature with a focus on
making students feel successful in developing positive and nurturing bonds with a
trusted adult on campus (Resnick et al., 1993; Schwartz, 1999; Sprague et al., 1998).
Wilson’s (2004) further examination of the interactions of school climate and
school connectedness found that despite the effects of school climate, strong school
connectedness is a significant protective factor. Both school climate and school
connectedness act almost independently in predicting aggression. In various levels of
school climates, including positive to negative climates, students with high
connectedness are less likely to experience high levels of victimization and
aggression. This implies that connectedness can ultimately alter the climate of the
school. For example, responsive teachers who are connected to their students are
more likely to take responsibility for school campuses (Astor, Meyer, & Behre,
1999). Astor et al. (1999) also found that caring teachers build personal relationships
with students which inadvertently address school climate in a way that serves to
decrease the fear of students who often avoid areas perceived as unsafe and
dangerous.
In further addressing school connectedness, Eisenberg, Neumark-Sztainer
and Perry (2003) found the following positive effects, “.. .[students] earn higher
grades are less likely to smoke cigarettes, use drugs, have an early sexual debut, be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
involved in violence, and be emotionally distressed than their less connected and less
successful peers” (p. 311). Although Eisenberg et al. reported responses of 7th-10th
grade student self reports, positive effects of younger children are supported by
O’Donnell, Hawkins and Abbott (1995). An initial survey of aggressive 10 and 11
year old boys followed with surveys during the two subsequent years indicates that
school and family bonding are protective factors against serious delinquent behavior
and substance abuse at age 13 and 14 (O’Donnell et al., 1995).
In addition to school and family connectedness, peer and community
connectedness are essential factors in violence prevention (Hawkins et al., 2000;
Potter et al., 2001; Smith & Sandhu, 2004). Research by Smith and Sandhu (2004)
indicates positive effects of academic achievement and feelings of happiness and
well being when students are bonded to these environmental factors. Positive
relationships between a parent or caretaker and child are based on the same theory of
connectedness. Schools have a significant role in building parent-school and parent-
child relationships by providing parent and family based preventative measures
where primary caretakers are taught communication and conflict resolution skills. In
addition, research finds that the earlier the intervention, the greater the influence
there is of reducing aggression and antisocial behaviors (Smith & Sandhu, 2004).
Family connectedness can also be strengthened by including parents in
prevention planning and strategizing. Family connectedness is defined by a child’s
feeling of closeness to family members and the feeling of being cared for (Resnick
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
et al., 1993). When parents are knowledgeable about the important issues and are
educated and made aware of basic warning signs of violence, they take ownership
and responsibility for deterring their child’s undesired aggressive behaviors. On the
other hand, children who view an entrusted adult or peer as unresponsive in tackling
problem behaviors such as bullying are more susceptible to higher levels of
victimization (Astor et al., 2003). This would be problematic in elementary schools
since most cases of bullying occur in the younger grades.
A high incidence of violence is also found in schools where students feel
isolated and socially withdrawn (Hunt et al., 2002; Warner, Weist, & Krulak, 1999).
Alienation occurs in climates that lack connectedness, leading to the assumption that
alienated students experience higher levels of violence. According to Hranitz and
Eddowes (as cited in Warner et al., 1999), aggression in students may result from
feeling alienated, disconnected and powerless.
In 1993, the Comer School Development Program of the Yale Child Study
Center in New Haven, Connecticut nationally surveyed 150 elementary, middle and
high school parents, staff and students about their perceptions of school violence
(Haynes, 1996). The survey resulted in only 33 percent of the students feeling that
the teachers can be trusted to protect them from violence. This data symbolizes a
general feeling of a lack of positive bonding with a trusted adult on campus.
Numerous other studies are based on self-report surveys completed by middle
and high school students encompassing 7th to 12th grades. One such report of self
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
perceptions of 7th -12th graders in Minnesota, shows the protective factor of school
connectedness as more prominent than family connectedness for acting out behaviors
(Resnick et al., 1993). Implications for these results place a heavy weight on schools
to provide an environment of caring and belonging. A second self report survey of
9th to 12th graders from eight public schools reported the positive relationship
between school connectedness to both improved academic performance and
increased participation in extracurricular activities (Bonny, Britto, Klistermann,
Homung, & Slap, 2000). In addition, Eisenberg et al. (2003) report a negative
correlation between peer harassment and school connectedness where middle and
high school students who reported liking school reported less incidents of peer
harassment.
The literature is replete with evidence of the relationships between both
school climate and connectedness as preventative measures against violence on
campus. Both factors have also been referred to as measures of prevention for
schools developing a more integrated and comprehensive solution to curb dangerous
behaviors of school children. Definitions and characteristics of positive school
climates have included the three other elements of a comprehensive conceptual
model of safe schools as illustrated in Figure 1. This section presented school climate
and connectedness as the foundation for building three other elements of safe
schools: high expectations, discipline and crisis planning, and a collaborative
community partnership. This next section will review the literature on the patterns
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
and effects of high expectations on school violence, followed by the need for
collaborative community partnerships and clear goal-driven discipline and crisis
plans.
High Expectations
The task of building skills of resiliency in students is often referred to in the
literature as a positive component of violence prevention programs. Christie,
Jolivette and Nelson (2000) define resiliency as, “the ability to recover strength and
spirit under adversity on both internal (self) and external (family, school,
community) factors for a positive outcome” (p. 3). A study by Finn and Rock (1997)
demonstrated the relationship found between resilient students and suspensions,
which is the indicator of school violence used in this study. Resilient students were
found to have lower numbers of in and out of school suspensions at 10th grade. These
results indicate that by building resiliency skills in students, there will be less
incidents of violence leading to suspensions (Finn & Rock, 1997).
Children who are labeled at-risk or who are surrounded by numerous other
school and family risk factors may have difficulty escaping from stereotypes that are
sometimes played out in the classrooms as teachers lower their expectations.
However, with the presence of protective factors such as resiliency, between 50-80
percent of these children have proven to elude the expected outcomes of aggression
and violent behaviors and break the cycle of self-fulfilling prophecies (Benard, 1993;
Benard, 1997; Christie et al., 2000).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
Resiliency research reports protective factors to include high academic and
social expectations, connectedness with teachers, peers and family and being
provided with opportunities for successful participation in school (Bernard, 1997;
Christie et al., 2000). This suggests that teachers and school staff in addition to
family members play an important role as models to children to provide such
protective factors in the classrooms and homes (Benard, 1993). High expectations for
both academic achievement and social behaviors have emerged as the third element
of safe schools.
High expectations for both academic and social behaviors have been
identified as a protective factor in successful schools where teacher expectations are
clear and students are involved in school (Benard, 1991). The Comer School
Development Program, a successful safe schools model of 150 elementary, middle
and high schools, includes high expectations for student achievement as one of their
12 components of a safe and secure school climate (Haynes, 1996). This is based on
the premise that students can learn when they are challenged, provided with adequate
resources and are perceived capable of completing tasks by school personnel
(Dwyer, 1998; Haynes, 1996).
The literature indicates that the expectancy of teachers affects students’ self
esteem, self-efficacy and academic achievement. Teachers’ stereotypical beliefs of
students affect how they interact with them and how the students begin to feel about
themselves (Haynes, 1996). The development of student discipline policies for an
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
inner-city elementary school resulted in eight proactive approaches that emerged to
address student discipline, classroom management and school violence. Discussions
were held in focus groups throughout the year, comprised of university faculty, pre
service teachers, social workers, school administrators and school staff and staff
from an involved community center. The suggested approaches to address discipline
policies included the following, holding high expectations for all students
implementing engaging and appropriate educational activities, coaching for self-
discipline, including instruction in conflict resolution skills, modeling appropriate
behaviors, encouraging home-school linkages, and supporting multisystem and
multisector community involvement (Ward, 1998, p.39). Participants communicated
that social skills and appropriate behavior can be learned with the expectation that all
students are able to take responsibility for their own behavior.
Lohrmann and Talerico (2004) found that clear and positively stated
behavioral expectations reduced disruptive classroom behavior. Ten fourth and fifth
graders were introduced to three teacher selected behaviors that were targeted in this
study. As a result of classroom time devoted to explanation, teacher demonstration of
expected behaviors and the implementation of positive reinforcement for observed
desired behaviors, students met teacher expectations. One caution of this study is the
effect of the positive reinforcement on the results and the factor of external
motivation on the modification of student behavior in this study. More research is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
needed to determine the effect of such a reward system in comparison and in
addition to setting high social and behavioral expectations for students.
Astor et al. (1999) interviewed school students, teachers and other school
staff from five American high-schools to measure the extent to which the most
serious incidents of victimization occurred in various locations and times. Students
reported the importance of relationships with a caring member of the school staff as
well as the importance of prompt responses to intervene in violent incidents between
students regardless of location (Astor et al., 1999). Caring teachers in this study were
defined as those who held high expectations on the quality of students’ work and
monitored students’ attendance. According to the participants in Astor’s et al. study,
the most effective interventions were the physical supervision of caring teachers who
made an effort to connect with students and the consistent application of violence
policies.
Collaborative Community Partnerships
Pollack and Sundermann (2001) stress the importance of following a
strategic process in the development of a comprehensive school prevention plan. The
literature on comprehensive prevention planning proposes starting with a
collaboration of stakeholders in education, including school staff, students, parents
and community members (Dwyer & Osher, 2000; Kadel, 1999; Pollack &
Sundermann, 2001).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
Kellam, Printz, and Sheley (2000) suggest a three step developmental process
in creating collaborative community partnerships. The first step includes establishing
school boards that consist of the local community and institutional leaders to
encourage their active participation in the decision making of prevention programs.
The second step of Kellam’s et al. plan to develop partnerships is to assess the values
and priorities of the members and population in order to recommend prevention
activities that meet the needs while best representing the values of the community.
The last step is a political one and involves negotiation with leaders on the board to
ultimately arrive at the satisfaction of all parties’ interests.
The three step process and implied purpose for the community boards seem
more self-serving for the various interest groups that comprise it than other
collaborations found in the literature (Kellam et al., 2000). Others, rather, state their
intended purpose of collaborative community partnerships to include sharing the
responsibility for protecting students, educating parents and community members of
the early warning signs of violent behavior as well as educating parents about child
development issues, and placing responsibility and appropriate consequences on
students for their actions (Dwyer et al., 1998).
Kadel, Watkins, Follman and Hammond (1999) suggest partnerships with
students, parents, other professional, law enforcement officers, community members,
and other agencies such as social workers, early childhood specialists and medical
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
practitioners. Other researchers support law enforcement models in their
comprehensive prevention collaborative teams (Dwyer et al., 1998; Sprague et al.,
1998).
In describing the major approaches and components of a violence-prevention
program, Peterson and Skiba (2001) listed parent and community involvement,
character education, violence-prevention and conflict resolution curriculum, peer
mediation, and bullying prevention. The parent and community involvement
component is the most recognizable of the components listed in a majority of
research about comprehensive prevention models for school safety. Peterson and
Skiba further imply indirect links between parent involvement and decreased levels
of violence on campus. The implication is that when parents are more actively
involved in their child’s education, the home environment will change to reflect the
values of the school while communication between the home and school improves.
Parent involvement has also been positively associated with increased levels of
academic achievement, which has then been identified with an inverse relationship to
aggression (Eisenberg et al., 2004; Peterson & Skiba, 2001; Wilson, 2004).
Another variation found in the literature among various comprehensive plans
for safe schools includes the number of suggested teams in a single school. While
some plans operate under one team, Dwyer and Osher (2000) suggest the formation
of two separate groups to address school safety, each consisting of at least three of its
members serving on both teams; the principal, a teacher and a mental health
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
specialist. Dwyer and Osher’s report, Safeguarding our children: An action guide, is
the follow-up from a previous report, Early Warning, Timely Response (Dwyer et al.,
1998). It is based on research that has established the value of community
collaborations in comprehensive plans of school safety. Dwyer and Osher (2000)
propose that the first team serves to evaluate the overall performance of the school,
including academic and behavioral achievements and school climate. Within this
group, community collaborations are developed to best access the talents and
community leaders as well as to address the community’s needs. The second team
assists individual students who exhibit early warning signs making the appropriate
referrals to the best fit resources. This team responds through early intervention with
active participation of the students and families in the referral process (Dwyer &
Osher, 2000).
A proven successful model of a comprehensive prevention plan that is based
on over 30 years of data and includes the use of collaborative community
partnerships is the School Development Program in New Haven, Connecticut
(Haynes & Comer, 1996). A study that began in 1968 in two elementary schools has
developed three decades later into a full-grown, evidence-based program (Comer &
Haynes, 1991). This program is composed of three teams: the school planning and
management team, the students and staff support team, and the parent program team.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
The first team is the leader of the three and serves the purpose of addressing and
improving the academic and social climate of the school. It consists of the principal,
teachers, other school staff, parents, students as well as community members.
The student and staff support team is the second in the School Development
Program Model (Haynes & Comer, 1996). Members with mental health training
comprise this team whose purpose is providing prevention and intervention measures
to prevent hazardous situations. Community relationships play an important part of
this team as community resources are best fitted with individual needs of each
student. Comer and Haynes (1991) also state that the results of this team’s
intervention are usually stronger bonds between the school, parents, students and the
community.
The third beneficial group in the School Development Program Model is the
parent program that encourages parent participation at three different levels (Haynes
& Comer, 1996). At the first level, parents serve on the school planning and
management team to plan and supervise school activities. With direct links to the
community, parents bring their knowledge about community concerns as well as
practical solutions to best fit the needs of the students and community (Comer &
Haynes, 1991). Parents at the second level, volunteer or become paid assistants in the
classrooms. These parents interact with the students in various needed areas on
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
school campus from participation in the classrooms to supervision in the hallways
(Haynes & Comer, 1996). Finally, parents at the third level attend parent group
meetings and are involved in school social events.
The differences between the two teams suggested by Dwyer and Osher
(2000) and three teams of Haynes and Comer (1996) and Comer and Haynes (1991)
are slight. While both comprehensive school plans address whole-school initiatives
to improve school climate and implement supportive community collaborations
(principals, teachers, other support staff, students, parents, and community members
and agencies) in addition to an intervention component to address individual student
behavioral needs, Dwyer and Osher’s plan lacks the parent team component.
Another successful model of a collaborative community partnership is the
Friedens Haus coalition in St. Louis, Missouri (Ward, 1998). A partnership existed
between an inner-city elementary school, a social service agency, a church and the
Hyde Park community with participation from approximately 50 agencies,
organizations and businesses. Focus groups were first held with participants of all
partners as a needs assessment and resulted in eight suggested proactive approaches.
Following these suggestions was the establishment of a discipline committee to
create and implement policies to improve student behavior and academic
achievement as a means to alleviate criminal behavior in a neighborhood plagued
with crime. Results of this collaborative community effort include increased SAT
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
scores, the highest attendance rate in years, lower number of discipline referrals, and
the highest number of parent volunteer hours in the system.
Once collaborative community partnerships are established and ready to act,
the literature suggests sequential steps be taken in the process of building an
effective comprehensive safe school plan. Those steps are outlined in the next
section in review of the last element of the comprehensive conceptual model of safe
schools as previously illustrated in Figure 1, disciplinary and crisis plans.
Response Planning: Discipline and Crisis Plans
For schools just embarking on the journey of creating a comprehensive
safety prevention plan, a sequential order of events is suggested by the literature.
Establishment of a collaborated school-community partnership is the first step. These
chronological steps help to strengthen the process as schools undertake school-wide
prevention strategizing. Pollack and Sundermann (2001) listed the five critical steps
in the process as, 1) creating a planning team, 2) conducting a comprehensive needs
assessment, 3) developing a comprehensive school plan to address target areas and to
set goals, 4) researching and identifying strategies and program models, and
5) evaluating the program.
These five steps discussed by Pollack and Sundermann (2001) are supported
by other researchers who review comprehensive prevention strategies and planning
(Boulter, 2004). Another important step that emerged in the literature was response
planning, which includes discipline and crisis management plans (Dwyer et al.,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
1998; Kadel et al., 1999; Skiba & Petterson, 2000; Sprague et al, 1998). The
components of response planning comprise the fifth and last element of the
comprehensive conceptual model of safe schools in this paper. This element
combined with brief descriptions of the selected steps leading to it is discussed
below.
Prior to the response planning in the school-wide undertaking of a
comprehensive prevention plan, a collaborative community partnership must be
developed. With the strong leadership of this school-community team, a needs
assessment is recommended to evaluate the scope of the problem and the areas of
highest need (Hunt et al., 2002; Pollack & Sundermann, 2001). A needs assessment
also serves the purpose of identifying current performance and efforts, refocusing the
school’s priorities (Pollack & Sundermann, 2001).
In reference to bullying in schools, Astor et al. (2003) note school-wide
awareness of the problems as a crucial element to prevention. Awareness is
necessary to make the effectual changes and address the real problems faced by
schools. Achieving this awareness necessitates a needs assessment of the school.
Methods of data collection include surveys, questionnaires, and interviews from
which school profiles can then be developed and used to identify areas of strengths
and weaknesses (Pollack & Sundermann, 2001). An evaluation of the perceptions of
teachers, students and parents is a necessary component of an effective prevention
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
program in order to gain a better understanding of where the serious problems seem
to be.
Astor et al. (1999) surveyed students and school staff about locations of
actual violent incidents and perceived unsafe locations on campus. A disparity
resulted among the responses of school personnel and students when asked about the
discipline measures carried out in school against students (Astor et al, 1999). School
personnel reported suspensions and expulsions as the most common response to
violent behavior while students felt that suspensions were unfair and ineffective as a
one-size-fits-all response to violence. Students suggested that staff consider
individual cases and other good qualities of the students (Astor et al., 1999).
Out-of-school suspensions are a common form of discipline and although
there are definite times where violent students must be removed from a situation,
research presents the limitation of suspensions. Kadel et al. (1999) lists several
disadvantages of suspension as a disciplinary response. Often times, out-of-school
suspensions are enforced out of convenience without realizing the implications of
releasing potentially dangerous children in the streets. Other disadvantages include
positive relationships between suspended children and failed courses, increased drop
out rates, the positive reinforcement of being removed from an environment where
they do not wish to be, increased delinquency, and a disproportionate number of
minorities in the category (Kadel et al., 1999).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
Kadel et al. suggest more personable approaches to disciplining violent
students. Some suggestions included the review of student records and active
communication with parents to expose the student’s history. Data learned from both
alternate approaches may reveal the best intervention method for the individual
student, whether it is seeking assistance from a mental health specialist or realizing
the existence of warning signs.
Skiba and Peterson (2000) report a positive relationship between suspensions
and drop out rates found in the High School and Beyond study. Suspensions may
also have a long lasting negative effect on antisocial students who already feel
alienated from teachers and peers (Skiba & Peterson, 2000). The consequence in this
case is the negative affect on bonds between school and troubled students.
Connectedness is essential, however, as a protective factor in the violence prevention
models. The work by both Astor et al. (1999) and Skiba and Peterson (2000) imply a
relationship between suspensions and school connectedness. They also imply a need
for the student voice in school decision making venues. In constructing disciplinary
plans, the comprehensive prevention team must consider the perceptions and
attitudes of all stakeholders in education, including principals, teachers, support staff,
students, parents and community members.
Morrison and Furlong (1994) make an insightful statement about student and
staff characteristics as one of the four safe school dimensions developed in
California through Safe Schools: A Planning Guide fo r Action. Students arrive at
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
schools with prior experiences that may either positively or negatively affect their
perceptions and school relations. School policy and procedures ultimately act as
protective factors which build resiliency in victims who were previously
overwhelmed by risk factors leading to enhanced “student productivity and
satisfaction” (Morrison & Furlong, 1994, p. 9).
This suggests the magnitude of developing school disciplinary policies that
are both clear and ingrained in all of the primary stakeholders of schools, including
the principals, teachers, other school staff, students, and parents. In his review of
elementary school data on violence, Astor (1995) lists considerations directed to
school personnel when creating school policies and undertaking a school-wide
approach to prevention and discipline. He makes the following points, 1) violent
children may be victims and perpetrators, 2) violent children may view adults as
unfair, inconsistent and ineffective at effectively responding to violence against
them, 3) physical harm is worse than psychological harm and this should be reflected
in the policies and consequences, and 4) positive integration of aggressive children
into school social settings may break cycles of victim-perpetrator and improve
connectedness felt by these children. All of these ideas should be considered by the
school-community team when developing effective discipline and crisis plans.
Traditional methods of handling problem behaviors are punitive and reactive
relying typically on detention, suspension and expulsion. However, there has been an
emergence of a more comprehensive preventative approach labeled by a few as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
“proactive school discipline plans” and “behavior management models” (Colvin &
Kameenui, 1993; Sprague et al., 1998; Ward, 1998). Sprague’s et al. (1998)
proactive school discipline plan identifies critical components of a safe school plan.
He attributes the success of a discipline plan to four essential components, which
include, 1) school-wide policies and procedures, 2) concerns of specific settings and
locations, 3) classroom management, and 4) assessment and support that are
individualized (Sprague et al., 1998).
Additionally, Colvin and Kame’enui (1993) studied Project PREPARE that
addresses the problems of existing school-wide discipline models. The premise of
this study is that of a proactive approach which empowers the student through
instruction to make the appropriate responses. Two middle schools participated in
this study, one received training of PREPARE while the other school comprised the
control group. The experimental school resulted in 50 percent fewer office referrals
and decreased office conferences, suspension, detention, and parent meetings.
Dwyer et al. (1998) further outline the essential steps in creating a violence
prevention plan. The first step is to delineate the roles and responsibilities of all
members of the school community to prepare them to recognize warning signs of
troubled students. Each member should be ready to act in necessary situations.
Suggestions for written plans include the following components, 1) descriptions of
early warning signs of probable dangerous behaviors, 2) descriptions of actions
carried out by school that address prevention, 3) descriptions of early and more
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
intensive intervention actions practiced by school, and 4) a crisis management plan
that includes both immediate responses to violence and a contingency plan after the
incident has occurred (Dwyer et al., 1998).
Kadel’s et al. (1999) model of discipline strategies include descriptions of
several options commonly used in schools: out-of-school suspensions, in-school-
suspensions and expulsions. In-school suspensions offer an alternative to out-of
school suspensions. An in-school suspension removes the violent or disruptive
student from the classroom and places him in another class where he can provide a
service to the school. It also removes the privileges of extracurricular school
activities such as athletics. Expulsion is the last step in disciplining students and
should apply only in the most serious situations.
Alternatives to suspensions are suggested in cases that are less serious in
nature (Kadel et al.,1999). Some schools offer the option of supervised service
assignments where they provide a service to the school such as removing graffiti
from hallways and tutoring younger students. An additional alternate to expulsions is
enrollment in an alternative educational program that may better serve the needs of
the troubled child.
Kadel’s et al. (1999) compilation of research and successful models from the
national, state and local levels are among the most detailed reports about creating a
crisis management response plan. Examples of crises include the use of weapons on
campus, bomb threats, physical fights, natural disasters and suicides (Dwyer et al.,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
1998). In any of these cases, the school community should be prepared to take fast
action in assessing the situation and acting appropriately thereafter. Both Kadel et al.
(1999) and Dwyer et al. (1998) agree on the following components of a crisis
contingency plan, 1) inservice training and practice of the plan in anticipated violent
situations, 2) a written manual or a list of individual responsibilities and duties of
each school community member, 3) proper and immediate communication with
outside resources such as law enforcement and other community agencies, 4) a
coordinated community response to aid victim recovery addressing both physical and
emotional needs, 5) counseling of victims, and 6) immediate re-entry to school
environment.
Additionally, descriptions of effective communication with a variety of key
players should be included in crisis plans (Kadel et al., 1999). First and foremost,
teachers must be aware of what is happening in order to best respond appropriately
to the situation. Serious events should be disclosed to teachers at staff meetings
immediately following the crisis. Effective communication should also be practiced
with parents about the situation, with immediate notice if a situation involves their
child. Students should be a part of the crisis planning and participate in practices of
anticipated situations. They should also be generally told, without releasing names of
the students involved, of violent episodes. Lastly, one staff member should be
assigned to communicate with the media as the official spokesperson. Other staff
members would then act to intercept and direct all media to the spokesperson.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
The reporting of violent incidents is also very important (Kadel et al.,1999).
This leads to continued updates of the school profile and empowers the school with
evidence to make the necessary changes to existing policies and procedures. Dwyer
et al. (1998) suggest that crisis management plans include evacuation procedures that
identify a safe place to gather and escape from imminent danger.
Of all the five elements of safe schools, the discipline and crisis plans is the
only interventionist component as the others are more preventative in nature. All five
elements of this study’s comprehensive conceptual model of safe schools are based
on research and successful models of safe schools. These five elements- school
climate, connectedness, high expectations, collaborative community partnerships and
discipline and crisis planning- all demonstrate a critical need to connect with
students in order to recognize early warning signs of potential violence. Additionally,
there is a need to carefully develop a comprehensive and systematic violence
preventive program to serve the specific needs of the individual school and
community. The next section will turn to the responsibility of providing safe schools
as indicated by research, perceptions of the key players in education and by the
courts with examples of legislation affecting violence.
Responsibility for Providing Safe Schools
Although most agree that violence in schools is a problem that desperately
needs attention, what is usually not agreed upon is the answer to the question, “Who
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
is ultimately responsible for school safety?” The responses, usually stemming from
perceptions of the key stakeholders in education, are common within groups but vary
greatly between them. The act of finger pointing between the groups is widespread
as shown throughout the literature (Leinhardt & Willert, 2002; Stetson, Stetson, &
Kelly, 1998; Thomerson, 2000; Willert & Lenhardt, 2003).
The following section begins with a review of the literature about
responsibility as reported by the perceptions of teachers, students, parents and
community agencies.
Responsibility based on research and key-member perceptions
Three main studies report the perceptions of middle and high school students,
school personnel, parents and community agency representatives on their views
about school violence and responsibility (Leinhardt & Willert, 2002; Stetson,
Stetson, & Kelly, 1998; Willert & Lenhardt, 2003). Lenhardt & Willert (2002)
interviewed these five stakeholder groups from 13 districts in New York and
responsibility for school safety emerged as one of the themes. Although, both
students and adults agreed that school violence is everyone’s responsibility, finger
pointing still occurred between the parents and school personnel.
According to Lenhardt & Willert (2002), school personnel felt that more
responsibility needs to be shifted back to parents who are often disconnected with
their own children. Parents, on the other hand, asked for more accountability and
effective communication from school personnel. Parents also felt disengaged and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
unwelcome in the schools. Similarly, community agencies felt unwelcome and
placed the blame on the schools for not taking the lead to . .establish a team
approach” (Lenhardt & Willert, 2002). Lastly, students reported a desire to be held
more responsible for school violence and also felt that their opinions were
unwelcome. In conclusion, Lenhardt and Willert (2002) recommended that schools
lead the movement towards establishing shared responsibility with the community.
In addition, schools should also bare the responsibility of providing more
opportunities for involvement.
In a later study, Willert and Lendardt (2003) sampled participants from the
same five stakeholder groups from 17 districts in New York. Results of this study are
similar to the first. Middle and high school students, parents, and representatives of
community agencies reported a disconnect with the schools asserting that schools do
not listen to them. Both parents and community agencies shared a desire to be more
involved as to share the responsibility with schools, however, they felt excluded
from the process. Both studies amplify the importance of school connectedness and
collaborative community involvement as protective factors in effective violence
prevention programs.
The third study by Stetson, Stetson and Kelly (1998) identified teachers’
perceptions of violence in society, schools and the responsibility of schools in
providing a safer society. Seventy-eight elementary, middle and high school teachers
completed the survey, Are School’ s Responsible fo r a Civil Society? (Stetson &
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
Stetson, 1998). Of the 78 teachers, 95 percent of them reported perceptions of
increased violence, while 84 percent felt that violence increased in their own schools.
Not surprisingly, 51 percent of the teachers attributed an increase in violence to
parental influence or a lack of parental involvement. In addition, 58 percent of the
teachers reported that schools are “extensively responsible” for safe societies, while
21 percent of the teachers responded either “quite responsible” or “some
responsible” regarding the extent of school responsibility. This study adds the piece
of perceived school responsibility of teachers that schools should be liable for
providing safe campuses. Regardless of assigning blame on the other stakeholder
groups, the consensus is that effective prevention programs develop collaborations
between school and community.
Two additional studies examined teacher perceptions that their roles and
responsibilities lie strictly in the classroom (Behre, Astor and Meyer, 2001; Furlong
and Morrision, 2000). Behre et al. (2001) further explored elementary and middle
school teacher perceptions of their reasoning to intervene in a physical fight at
school. Surveys and follow up interviews with 54 elementary and 54 middle school
teachers revealed that 17.9 percent of elementary teachers and 57.5 percent of middle
school teachers would respond differently in their classroom than in the hall.
Although most elementary teachers responded that they would intervene in fights in
the halls to stop a student from physical harm, middle school teachers responded
with multi-faceted reasoning such as a feeling of efficacy in those undefined areas
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
and a lack of personal responsibility. A decreased sense of personal responsibility in
undefined areas on campus accounted for 13.6 percent of teachers’ reasoning that
they would act differently in places other than their classrooms.
Furlong and Morrison (2000) refer to the term “owning school violence.”
They state, “If one embraces a school violence rather than a violence that happens in
schools definition, attention may be refocused on the role that school as a physical,
educational, and social environment plays in violence among its participants” (p. 74).
Statements such as this are made to encourage schools to take more responsibility for
keeping their schools and thus their students safe from harm. Furthermore, Furlong
and Morrison (2000) state, “Threat of physical harm can be interpreted additionally
as threat of developmental harm; that is, the threat of reality of physical harm has
consequences that suppress the maximal educational growth and development of
students” (p. 74).
Federal guidelines set under the No Child Left Behind Act of 2001 and the
Hawai’i Department of Education indicate a concern in the area of school safety and
drug free schools nationwide. Even after violence and drug prevention programs
have been established within the schools and both the students and staff are educated
about right and wrongful actions, violence between students still occur. Teacher
perceptions of responsibility play a vital role in understanding their decisions to
intervene or not while identifying areas of need for strengthening school partnerships
with the Department of Education, the home and the community.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
Conclusion
Although national and state incidence rates of violence in schools have
decreased over the past decade, violence continues to pervade schools throughout the
nation. Numerous studies have explored risk and protective factors of violence in
adolescence and drive the value for a more multi-faceted comprehensive and
systemic prevention plan. Throughout the literature, the basic elements included in
this study’s comprehensive conceptual model of safe schools have emerged as
separate units or in combination of several elements. This study has explored a more
preventative comprehensive approach to addressing violence in elementary schools
and is inclusively comprised of all five of what has developed as safe school
elements for this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
CHAPTER 3
RESEARCH METHODOLOGY
Introduction
The purpose of this study was to examine the perceptions of elementary
school teachers about school violence in O’ahu, Hawai’i. This study also
investigated the perceived existence of the five elements of safe schools across
schools with low, moderate and high violence levels as determined by each school’s
suspension rates. Lastly, this study examined the perceived responsibility for
providing safe schools among all stakeholders in education.
This study exists as one of three research projects that examined teacher
perceptions of school violence, the existence of safe school elements and a
comparison between schools across three levels of school violence. The other two
research projects examined perceptions in the middle and high schools in O’ahu,
Hawai’i. This chapter includes the research questions, the hypotheses, and a
description of the research methodology. The section on research methodology
includes the sampling procedure and population, instrumentation, and procedures for
data collection and analysis.
Research Questions and Hypotheses
Perceptions o f School Violence
Research Question 1: What are teachers’ perceptions of observed risk
behaviors, violent behaviors in school, and overall magnitude of school violence in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
elementary schools on O’ahu? Are there differences in perceptions of observed risk
behaviors, violent behaviors in school, and overall magnitude of school violence by
gender, years of teaching experience, years of teaching at school site, or violence
level?
Research Question la: What are teachers’ perceptions of violent behaviors in
school, and overall magnitude of school violence in O’ahu elementary schools?
Research Question lb. Are there differences in the perceptions of violent
behaviors at school by gender, years of experience, years at particular school site or
level of violence?
Hypothesis lb: Females will report a greater frequency of violent behaviors
at school than males. Less experienced teachers will report greater frequency of
violent behaviors than veteran teachers. Less experienced teachers at a particular
school site will report greater frequency of violent behaviors than veteran teachers at
that same site. With increased school violence levels, greater occurrences of violent
behaviors will be reported.
Perceived Existence o f Safe School Elements
Research Question 2: Is there a significant difference in elementary school
teachers’ perceptions of the five safe school elements (school climate,
connectedness, expectations, collaborative community network, and discipline/crisis
plans) between teachers from low violence, moderate violence, and high violence
schools on O’ahu?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
Research Question 2a: Is there a difference in school climate scores for
teachers from low, moderate, and high violence schools?
Hypothesis 2a: As the school violence level increases, less positive school
climate attributes will be reported along with more negative school climate
attributes.
Research question 2b: Is there a difference in connectedness scores for
elementary school teachers from low, moderate, and high violence schools?
Hypothesis 2b: As the level of violence increases, teacher connectedness
scale scores will decrease.
Research question 2c: Is there a difference in expectations scores for
elementary school teachers from low, moderate, and high violence schools?
Hypothesis 2c: As the level of violence increases, teacher expectations scale
scores will decrease.
Research question 2d: Is there a difference in collaborative community
networking scores for elementary school teachers from low, moderate, and high
violence schools?
Hypothesis 2d: As the level of school violence increases, teacher
collaborative community networking scale scores will decrease.
Research question 2e: Is there a difference in discipline and crisis plan scores
for elementary school teachers from low, moderate, and high violence schools?
Hypothesis 2e: As the level of school violence increases, teacher discipline
and crisis plans scale scores will decrease.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
Perceived Responsibility for Safe Schools:
Research question 3: To what degree do elementary school teachers perceive
the stakeholders (administrator, teachers, parents, students, community, resource
teachers and complex area supervisors, and DOE state level supervisors/personnel)
responsible for contributing to safe schools?
Research question 3a: How does perceived responsibility for providing safe
schools vary amongst schools with low, moderate, and high reported violent
incidents?
Hypothesis 3a: As the school violence levels increases, teachers report lower
perceived responsibility for each of the stakeholders (administrator, teachers,
parents, students, community, resource teachers and complex area supervisors, and
DOE state level supervisors/personnel).
Research Design
The research design of this study was a quantitative, non-experimental
comparative study. The schools that varied in reported levels of violence were
compared to further study the perceived differences between teachers teaching at
those sites as well as the relationship between the perceived existence of the five safe
school elements and violence levels. Overall perceptions of violent behaviors, and
observed risk behaviors in the schools were also studied. The extent of these
differences were measured by such factors as gender, total years of teaching
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
experience, and length of employment at particular school site, and the level of
school violence. Questionnaires created for teachers were the instrumentation in this
study.
Independent Variables
Gender. Gender was asked in question 1 of the Survey o f School Violence
and Safe School Elements (SSVSSE), the questionnaire used in this study (Collier,
Kelling, & Siaosi, 2005). Reponses included “male” and “female.” Males were
coded 1 and females as 0.
Years o f teaching experience. Teachers were asked about their years of
teaching experience in question 2 of the questionnaire. Responses and codes
included: 1=2 or less, 2=2-3, 3= 4-5, 4= 6-10, 5= 11=20, 6= 20 or more.
Length o f employment at school site. Question 3 asked teachers about their
years of employment at that particular school site. The responses included the
following choices: 1=2 or less, 2=2-3, 3= 4-5, 4= 6-10, 5= 11=20, 6= 20 or more.
Degree o f behavior problems. The reported level of school violence for each
school was determined by the percentage of the total student population suspended
during the 2003-2004 school year. Schools were then divided into three groupings
according to their suspension rates: “low,” “moderate” or “high” violence.
School climate. School climate was measured by question 11. Provided with
12 sub-questions, teachers were asked to indicate how well each described their
school with the following Likert categories: 1= strongly disagree, 2= disagree,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
3= agree, and 4= strongly agree. An exploratory factor analysis followed by a
confirmatory factor analysis determined two School Climate Scales— positive school
climate attributes and negative school climate attributes. Responses of sub-questions
for each scale were used to create two reliable scale scores. Six questions comprised
the positive school climate scale and individual scores ranged from 6-24. The
negative school climate scale scores ranged from 3-24 and also included 6 sub
questions.
Connectedness. Connectedness was measured by question 12 that included
nine sub-questions. Teachers were asked for the extent that given behaviors occurred
at their school with Likert scale responses that included “strongly disagree,”
“disagree,” “agree” and “strongly agree.” A factor analysis determined one reliable
scale from which a Total Connectedness scale score was created, ranging from 9-36
for each respondent.
Expectations. Expectations were measured in question 13 that asked teachers
to respond to seven sub-questions about the extent that each given behavior occurs at
their school. The four-part Likert scale was utilized, ranging from “strongly
disagree” to “strongly agree”. Responses from these questions were used to create an
Expectations scale score with a range from 7-28 for each teacher.
Collaborative community networking. Collaborative community networking
was measured by question 14 of the SSVSSE. Teachers were asked to evaluate their
relationships with each of the five provided groups including, PTA, support staff,
resource teachers and complex area supervisors, DOE state level
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
supervisors/personnel, and police. All groups act in some capacity as community
members to deal with and address school violence. Twenty-two sub-questions were
asked between the five groups and the following Likert categories were provided:
l=not at all, 2= a little, 3= some, 4= quite a bit, and 5= very much. A Collaborative
Community scale score was created that ranged from 22-110 for each respondent.
Discipline and crisis plans. Perceptions of the existence of discipline and
crisis plans were measured by question 15 on the teacher questionnaire. Provided
with seven sub-questions, teachers were asked to respond to the extent that provided
behaviors happened at their schools to address discipline plans with a four-part
Likert scale ranging from “strongly disagree” to “strongly agree”. Responses from
these seven questions were subjected to an exploratory and confirmatory factor
analysis from which a reliable Total Discipline scale was created with scores ranging
from 7-28 for each respondent.
Perceived responsibility for school safety. The perceived responsibility for
ensuring safe schools was measured by question 25 on the teacher survey, SSVSSE.
Teachers were asked about their perceptions of how much responsibility they placed
on a given list of stakeholders in education. Stakeholders included school, parents,
students, community (government/private agencies, law enforcement, etc...),
resource teachers and complex area supervisors, and department of education state
level supervisors/personnel. Responses were measured with the following 5-part
Likert categories: l=none, 2= a little, 3= some, 4= quite a bit, and 5= very much.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
Dependent Variables
Perceptions o f School Violence. The perceptions of school violence in
research question 1 were measured by an analysis of three questions on the teacher
survey. The first was the Observed Risk Behaviors questions included as question 7
on the teacher questionnaire. This question asked “To what extent do the following
behaviors happen at your school?” It consisted of 14 sub-questions that asked about
specific violent actions observed on campus. Teachers were asked to respond
according to the following 5-part Likert scale: 1= not at all, 2= a little, 3= some,
4= quite a bit and 5= very much.
Due to numerous non-responses from teacher respondents, factor analysis
was not appropriate. Rather, an Observed Risk Behaviors scale score was developed
for each respondent, ranging from 14-70.
The second dependent variable measuring perceptions of school violence was
the Violent Behaviors in School question in the teacher questionnaire (question 9).
This question asked teachers to estimate the number of incidents that had occurred
since the beginning of the school year. It consisted of 13 sub-questions that asked
about specific violent actions that happened in school, on the way to school, during
recess, or during classes. The following 5-part scale was used: 1= 0, 2= 1, 3= 2,
4= 3-5 and 5= 6+.
The last dependent variable measuring teacher perceptions asked about the
perceived magnitude of the school violence problem in the school. It measured
perceptions with the following 5-part Likert scale: 1= a very little problem or not a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
problem at all, 2= a very small problem, 3= a medium-level problem, 4= a large
problem, and 5= a very large problem.
Degree o f School Violence. The dependent variable for research question 2
was the degree of school violence on campus as measured by the percentage of total
student population suspended for the 2003-2004 school year. All O’ahu elementary
schools were divided into three groupings (low, moderate and high) according to
their suspension rates. Through a stratified random sampling procedure, one school
from each group was initially chosen. The results included three schools with the
following continuous codes: 1= low, 2= moderate and 3= high. A small sample size
for the moderate violence school resulted in the inclusion of three schools to
represent this particular sample.
Population and Sample
The population for this study was represented by teachers from a total of 121
O’ahu Elementary schools. The sample was chosen using a stratified sample
procedure. Schools were placed in an ordered array from schools with the least to the
greatest percentage of total student body suspended during the 2003-2004 school
year. The ordered array illustrated a representation of tri-modal groupings and the
schools were separated according to three groups. The first group comprised the
schools reporting low violent incidents (0%), the second group comprised schools
that report moderate levels of violent incidents (0 %> n < 1%) and the last group
represented schools with high reported violent incidents (1% < n < 6%).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
Unlike the other two parallel research projects that have focused on O’ahu
middle and high schools, quartiles were not practical in this study due to the majority
of the schools with reports of zero percent suspension rates for the year. In this case,
these schools crossed into the second and third quadrants making it difficult to define
the break between low and moderate levels of school violence. Subsequent selections
for two other moderate violence schools were necessary to meet the minimum
sample size requirements.
Each school was first assigned a school identification number from which
Microsoft Excel randomly selected one school from each of the three groups. The
school identification numbers were then matched back to the individual school sites.
The moderate violence schools consisted of three schools rather than a single sample
due to low response rates. Therefore, the sample included a total of 5 elementary
schools.
The number of suspensions was used in this study as a measure of low,
moderate and high reported violent incident schools because of its availability and
accessibility across O’ahu schools. Chapter 19: Hawai’i State Administration Rules
that includes Student Misconduct, Discipline, School Searches and Seizures,
Reporting Offenses, Police Interviews and Arrests, and Restitution for Vandalism
and Negligence delineates specific disciplinary guidelines for the State of Hawai’i
public schools (Office of Curriculum, Instruction and Student Support/Student
Support Services Branch Department of Education, 2002). Definitions of student
conduct deserving of suspensions are provided. The assumption made in this study is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
that reports across all schools in Hawai’i are similar and suspensions are based on
the same definitions for prohibited student behavior. Principals and other school
administrators who ultimately make the decision to suspend base their decisions on
clearly stated statewide guidelines. However, the presence of individual
interpretations of these statewide guidelines would be a possible limitation of this
study.
Demographics for Sample Schools
O’ahu consists of nine complexes under the Hawai’i State Department of
Education. Each complex consists of 2-4 high schools and their elementary and
middle feeder schools. The five randomly selected schools are located in three
different complexes. The reported demographics from the 2003-2004 school year of
the five sample schools are included in Table 2. Averages were taken and reported
for the three schools that represented the moderate violence schools. Although
teachers comprised the sample of this study, student demographics are included in
this table to reveal the community in which the teachers work.
Table 2 reveals the differences between the three school violence levels. The
schools varied significantly in the size of the student body during the 2003-2004
year, with the high violence school comprised of the largest student body followed
by the low and the moderate violence schools. Appropriately, the number of full time
equivalent teachers corresponds to student body size. The student to teacher ratio is
also highest for the high violence school with the greatest number of students and
lowest for the lowest violence school with the next largest student population.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
One notable difference between the sample schools was the percentage of
students receiving free or reduced lunch. The low violence school consisted of 8.3
percent of the student body receiving free or reduced lunch in comparison to 30
Table 2
Demographic Data o f Five Sample Schools
Demographic Data 2003-2004
Low
Violence
School
Moderate
Violence
School
(ave.)
High
Violence
School
School Data
Fall 2003 Student Enrollment 588 491 812
% of students suspended 0 0.68 3.2
Average daily attendance (% students
present) 96.9 94.4 94.5
% of students receiving free or reduced
lunch 8.3 30 73.8
% of students in special education 7.1 6.9 9.5
% of students with limited English
proficiency 9.7 2.13 14.8
% of 5th grade students proficient on
statewide reading assessment 69 50.7 45
% of 5th grade students proficient on
statewide mathematics assessment 45 25.3 14
Community Data
% of households in school district
receiving public assistance income 7.6 6.2 10.9
% of families with children living in poverty 11.2 7.9 9.4
Teacher/Staff Data
Total administrators (Principal, Vice
Principals, Student Activity Coordinators,
Student Services Coordinators,
Registrars, and Athletic Directors) 3 3 3
Total full-time equivalent teachers 30 30.5 52.5
Student to teacher ratio 19.6 15.7 17.3
Teachers' average years of experience 13.6 10.5 11.5
% of teachers with 5 or more years at
school 53.3 53.5 64.8
% of teachers with an advanced degree 30 27.9 19
% of teachers fully licensed 96.7 83.3 84.9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
percent in the moderate violence schools and 73.8 percent in the high violence
school. Another difference was in the percentage of teachers with advanced degrees.
A higher percentage of teachers from the low violence school had an advanced
degree (30%) followed by teachers from moderate violence schools (27.9%) and
high violence schools (19%). Similarly, the vast majority of teachers from the low
violence school were fully-licensed (low violence school, 96.7%; 83.3% at the
moderate violence school; and 84.9% at the high violence school).
Teachers across the sample were similar in several ways. First, fifty percent
or more of teachers at each school had taught at their site for five or more years.
Second, the daily average attendance rate for each school violence level was above
94%. Last, each school violence level consisted of three administrators to lead the
schools.
Instrumentation
A structured questionnaire, The Survey o f School Violence and Safe School
Elements (SSVSSE), was the instrument in this study (Collier et al., 2005). The
questionnaire consisted of questions covering the following topics: demographics,
perceptions of observed risk behaviors, perceptions of violent behaviors at school,
perceptions of the existence of safe school elements, teacher preparation and
preparedness for handling school violence, perceptions of prevention efforts and
their effectiveness and perceptions of school safety responsibility. The questionnaire
consisted of 27 multi-part questions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
The Survey o f School Violence and Safe School Elements (SSVSSE)
instrument used in this study is an adaptation of earlier research studies. One of the
main sources for this study’s instrument is the principal and homeroom teacher
surveys conducted in a national study of violence in Israel (Benbenishty & Astor,
2005). These survey versions were adapted to comprise the following sections in the
SSVSSE, Observed Risk Behaviors, Severity o f the School Violence Problem,
Prevention o f School Violence, and Collaborative Community Networking.
The instrument for this study was also an adaptation of the California School
Climate and Safety Survey (Furlong & Morrison, 1998). This instrument contributed
to the following areas of the SSVSSE, School Climate, and Connectedness. The 2001
Omnibus Safe School Survey (Henson, Massey & Armstrong, 2002) for students was
adapted and included in the Disciplinary Policies section of this study’s
questionnaire. A section of the 2003-2004 Principal School Survey on Crime and
Safety Questionnaire (U.S. Department of Education, 2003) was adapted and
included in the Crisis Plans section of the SSVSSE. Bernard (1991) and Skiba and
Peterson (2003) contributed to the section of Expectations on the instrument for this
study. Lastly, questions were adapted from Stetson, Kelly and Stetson (1999) and
comprised the section on Responsibility in the SSVSSE.
A pilot questionnaire was created and tested for additional reliability and
validity. Five veteran teachers, who were not participants in the final sampling, were
asked to complete the questionnaire. The results of the pilot survey were discussed
with the respondents and improvements were made to better extract the responses
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
necessary for further analysis. Such improvements include aesthetic changes of the
presentation of the tables, and changes in wording and phrasing to add clarity and
consistency to the responses. Suggestions were also made in several instances to
keep the response choices the same where applicable. Based on teacher responses to
the pilot questionnaire, a final questionnaire was developed.
Data Collection
This study analyzed data collected from the teacher questionnaires.
Permission to gather data from the sample was first obtained from the
Superintendent of the Hawai’i State Department of Education and the administrators
at the five selected sample schools.
Initial contact was made with the principal at each school, upon approval to
begin the data collection process, to further discuss the study and to arrange the most
opportune time to meet with the teaching staff. Questionnaires were distributed by
the researcher at school faculty meetings at the low and high violence level schools.
All three of the moderate violence schools utilized another distribution method. Due
to scheduling conflicts and a lack of time for a full presentation at a faculty meeting,
the administrators at these three schools distributed the surveys to the teachers
through either school mail or at a teacher meeting. Surveys were then collected at a
later time. The data were entered and analyzed using the statistical software program
SPSS.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
Data Analysis
The returned teacher surveys were analyzed through the statistical software
program SPSS with outputs of descriptive statistics to report the sample size,
minimum and maximum data, the mean and the standard deviation of the responses
of the independent variables. The data were then analyzed appropriately to carefully
address each of the three research questions. The detailed data analysis procedures
are provided in the next chapter along with the results.
Limitations
A limitation of this study is the method of determining the three levels of
violent schools. Although suspension rates were the most practical data to use due to
accessibility and availability, assumptions are made that these data are accurately
recorded and reflect the levels of violence on campus. Additionally, this study was
limited by the random selection of one school to represent the low and high violence
schools violence. On the other hand, results from the three schools that represented
the moderate violence level also present a limitation. Results may not accurately
represent violent occurrences within moderate violence level schools without
accounting for differences between those schools.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
CHAPTER4
PRESENTATION OF FINDINGS
Introduction
The purpose of this study was to examine the extent of school violence in
Hawai’i’s public elementary schools through self-reports of teachers in education.
Teacher perceptions of violence were compared by gender, years of teaching
experience, number of years at the present school, and level of violence present at
the school. Perceptions of responsibility were also explored. This chapter includes a
description of sample characteristics, the research questions, hypotheses, and the
results of the data analysis.
Sample
The final sample was comprised of 102 respondents from five elementary
schools. The response rates were as follows, 89 percent for the low violence school,
33 percent from the moderate violence schools and 77 percent for the high violence
school. Responses from three schools were combined to meet the moderate violence
school target sample size, while responses from single schools met the requirements
for both the low and high violence schools. In all three moderate violence schools,
questionnaires were handed out by the administrator and collected at a later time due
to time constraints at faculty meetings, which may have contributed to the low
response rates. Respondent characteristics are included in Table 3.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
Table 3
Teacher Response Rates by Gender, Years o f Experience, Years at School Site, and
School Violence Level
N
Percent of
Total Sample
Gender
Male 18 17.6
Female 84 82.4
Years experience
2 or less
9 8.9
2 - 3
3 3
4 - 5
6 5.9
6 - 1 0
31 30.7
1 1 -2 0
28 27.7
20 or more
24 23.8
Years at site
2 or less
19 18.8
2 - 3
9 8.9
4 - 5
12 11.9
6 - 1 0
25 24.8
1 1 -2 0
30 29.7
20 or more
6 5.9
Violence group
Low
32 31.37
Moderate
40 39.22
High 30 29.41
Overall 102 100
Research Questions, Hypotheses, and Results
Overall Perceptions o f School Violence
Research Question 1: What are teachers’ perceptions of observed risk
behaviors, violent behaviors in school, and overall magnitude of school violence in
elementary schools on O’ahu? Are there differences in perceptions of observed risk
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
behaviors, violent behaviors in school, and overall magnitude of school violence by
gender, years of teaching experience, years of teaching at school site, or violence
level?
Research Question la: What are teachers’ perceptions of violent behaviors in
school, and overall magnitude of school violence in O’ahu elementary schools?
Results—Observed risk behaviors. Three survey questions were reviewed to
determine teacher perceptions of school violence on campus. Question 7 in the
Survey o f School Violence and Safe School Elements (SSVSSE) asked respondents
about observed risk behaviors that happen in their school. Table 4 summarizes the
results of the observed risk behavior question.
It is important to note the high perceptions of bullying that occurred in the
five participating elementary schools (37.6 %-some, 7.9%- quite a bit), which is the
highest incident type reported by teachers. Vandalism and stealing were also quite
high in addition to perceptions that students get into fights.
Results— Violent behaviors in school. Second, participants were asked
to estimate the amount of times that a violent incident occurred on campus, on the
way to school, during recess, or during class since the beginning of the school year,
approximately during a four month range. This question in comparison to the
previous observed risk behaviors question encompassed a larger range of potentially
violent times throughout the day. Table 5 summarizes the results from this question
(question 9) on the Survey o f School Violence and Safe School Elements (SSVSSE).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
Table 4
Percentage o f Teachers ’ Perceptions o f Observed Risk Behaviors in School
Observed Risk Behaviors (Q7) Not at
All
A little Some Quite a
bit
Very
Much
Students get into fights
6.9 62.7 28.7 1.0 0.0
Students drink alcohol in school
93.1 6.9
0.0
0.0 0.0
Students destroy things (vandalism)
7.8 45.1 38.2 4.9 2.0
Students steal things from other students
or teachers
0.8
61.8 28.4 2.0 0.0
Students use drugs in school
92.2 7.8
0.0 0.0
0.0
Students threaten or bully other students
or teachers 11.9 41.6 37.6 7.9 1.0
Students bring guns to school
93.9 6.1 0.0 0.0 0.0
Students bring other weapons to school
(such as knives, sticks, etc.) 65.0 32.0 3.0 0.0 0.0
Overall, there is a violent atmosphere in
school 72.5 22.5 4.9 0.0 0.0
Outsiders (adults) enter the campus
during the school day and threaten. 73.5 23.5 1.0
2.0
0.0
There is dangerous gang activity in my
school 92.2 7.8 0.0 0.0 0.0
Teachers or other staff members come
on to, sexually harass, or bother students 0.0 0.0 0.0 0.0 0.0
Teachers or other staff members hurt
students verbally (insult, humiliate, and 63.7 32.4 3.9 0.0 0.0
curse)
Teachers or other staff members hurt
students physically (slap, pinch, push) 94.1 4.9 0.0 1.0 0.0
The 13 items of the violent behaviors in school question were subjected to
principal components analysis (PCA) using SPSS. Prior to performing PCA, the
suitability of data for factor analysis was assessed. Inspection of the correlation
matrix revealed the presence of numerous coefficients of 0.3 and above. The Kaiser-
Meyer-Oklin value was 0.824, exceeding the recommended value of 0.6 and the
Bartlett’s Test of Sphericity reached statistical significance, supporting the
factorability of the correlation matrix.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
Principal components analysis revealed the presence of three components
with Eigen values exceeding 1, explaining 39.71 percent, 16.72 percent, and 9.64
percent of the variance respectively. An inspection of the screeplot revealed a clear
break after the second component and it was therefore decided to retain the two
components for further investigation. These two components were retained for
further investigation, and supported by the results of a Varimax rotation. The rotated
solution revealed strong loadings on two components and all variables loading
substantially on only one component. The two components solution explained a total
of 56.42 percent of the variance, with Component 1 (Assaults, theft, and verbal
abuse) contributing 37.43 percent and Component 2 (Involving weapons)
contributing to 18.99 percent.
A confirmatory analysis verified the existence of two scales. The first
includes the assaults, theft and verbal abuse scale with a Cronbach’s alpha reliability
statistic of 0.905 and the second includes the violent behaviors involving weapons
scale with a Cronbach’s alpha reliability of 0.744, indicating reliable scales. To
protect the reliability of the scales, two sub-questions were removed to yield a higher
alpha statistic. These were questions were, “A student intentionally cut another
student with a knife or a sharp instrument,” and “Gang members threatened to hurt or
hit another student.”
Violent behaviors involving weapons in school were barely reported with 5
percent of respondents reporting a student who came to school with a gun at least
once since the start of school year. However small, any reports of guns and other
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
Table 5
Teacher Perceptions o f Violent Behaviors in School (percent)
Violent Behaviors in School (Q9) 0 1 2 3-5 6+
Assault, theft, verbal abuse
A student kicked or punched
another in order to hurt him/her 24.2 29.3 23.2 17.2 6.1
Student mocked, insulted, or
humiliated another student 14.9 26.7 19.8 13.9 24.8
(verbally)
A student threatened to hurt or hit
another student 26.0 27.0 20.0 18.0 9.0
Students ostracized (socially
isolated) a student 32.0 30.0 11.0 12.0 15.0
A student forcefully took things
from another student 46.0 29.0 12.0 10.0 3.0
Student stole personal belongings
or equipment from another 28.7 31.7 15.8 15.8 7.9
student
Student jumped on a student and
hurt him/her 37.6 28.7 15.8 13.9 3.0
There was conflict between
students from different ethnic 71.0 17.0 4.0 7.0 1.0
groups
Involving Weavons
A student threatened another with
a gun 99.0 1.0 0.0 0.0 0.0
A student threatened another with
a knife 96.0 4.0 0.0 0.0 0.0
A student came to school with a 95.0 5.0 0.0 0.0 0.0
gun
weapons on campus is cause for concern and warrants further investigation.
Incidents occurring six times or more for assaults, thefts and verbal abuse resulted in
the following percentages, mocking and insulting (24.8%), threats to hurt another
(9.1%), social isolation of student (15.0%), and theft (7.9%).
Results—Magnitude o f violence problem. The third question (question 8) in
the Survey o f School Violence and Safe School Elements (SSVSSE) asked for teacher
perceptions of the magnitude of the school violence problem in their school. This
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84
Likert scale question included the following responses: “a very small problem or not
a problem at all,” “a small problem,” “a medium-level problem,” “a large problem,”
and “a very large problem.” The majority of teachers perceived violence as a “very
small problem or not a problem at all” (70.6%), while 25.5 percent perceived
violence as “a small problem” and 3.9 percent perceived violence as a “medium-
level problem.” Table 6 summarizes the results of teacher responses and the
differences between gender, years of experience, years of experience at particular
school site and the level of school violence.
Table 6 demonstrates that most teachers viewed violence as a “very small
problem or not a problem at all,” although more males (77.8%) than females (69.0%)
agreed with this estimate. We do not know if the differences between male and
female teachers’ perceptions were significant as a chi-square analysis was not
appropriate here due to the violation of the assumption that 80 percent of the cells
should have expected frequencies of 5 or more (Pallant, 2005).
As expected, more teachers from the low violence schools perceived the
magnitudes of violence at their school as a very small problem or no problem at all
(87.5%) or a small problem (12.5%) in comparison to the moderate and high level
violence schools. Again, we do not know if these differences are significant because
of the problem with the cell size, rending the chi-square analysis inappropriate.
Research Question lb. Are there differences in the perceptions of violent
behaviors at school by gender, years of experience, years at particular school site or
level of violence?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
Table 6
Percentage o f Teachers ’ Perceived Magnitude o f School Violence by Gender, Years
o f Experience, Years at Particular School Site, and Level o f School Violence
A very small
problem or
not a problem
at all
A small
problem
A medium-
level problem
A large
problem
A very large
problem
Gender
Male 77.8 22.2 0 0 0
Female 69.0 26.2 4.8 0 0
Years of
experience
2 or less 66.7 33.3 0 0 0
2-3 100 0 0 0 0
4-5 83.3 0 16.7 0 0
6-10 80.6 19.4 0 0 0
11-20 60.7 32.1 7.1 0 0
20 or more 62.5 33.3 4.2 0 0
Years at
Particular site
2 or less 78.9 21.1 0 0 0
2-3 88.9 11.1 0 0 0
4-5 58.3 33.3 8.3 0 0
6-10 76.0 16.0 8.0 0 0
11-20 60.0 36.7 3.3 0 0
20 or more 66.7 33.3 0 0 0
Level of
School
Violence
Low 87.5 12.5 0 0 0
Moderate 63.3 30.0 6.7 0 0
High 62.5 32.5 5.0 0 0
Overall 70.6 25.5 3.9 0 0
Hypothesis lb: Females will report a greater frequency of violent behaviors
at school than males. Less experienced teachers will report greater frequency of
violent behaviors than the more experienced teachers. Less experienced teachers at a
particular school site will report greater frequency of violent behaviors than the more
experienced teachers at that same site. As the school violence level increases, greater
occurrences of violent behaviors will be reported.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
In determining the extent that teacher perceptions of violent behaviors in
school vary by gender, total years of teaching experience, length of employment at
that particular school site and the level of school violence at site, descriptive
statistics was first analyzed and compared. A factor analysis was completed to
determine reliable scales from which scale scores were reported for each respondent.
The reliability of the scales were then analyzed and confirmed with a Cronbach alpha
coefficient above 0.7 (Pallant, 2005). Items with the smallest distributions were
eliminated to keep the integrity of the internal consistency.
An independent T-test was first used to determine significance of scale scores
between males and females. Second, a one-way between groups analysis of variance
was conducted to determine existing significant difference in means between for
each of the independent variables tested (total years of teaching experience, length of
employment at that particular school site, and the level of school violence at site).
The Tukey-Kramer multiple comparison procedure further revealed which variables
significantly differed.
Results— Gender. An independent samples T-test was conducted to compare
Violent Behaviors in School- assaults, thefts and verbal abuse scale scores and
Violent Behaviors in School scale scores— involving weapons for males and females.
Table 7 summarizes the findings of the comparison of the means between males and
females for the assaults, thefts and verbal abuse scale scores, while Table 8
summarizes the findings for the violent behaviors involving weapons scale scores.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
Table 7
Independent T-Test for Assaults, Thefts and Verbal Abuse Scale Score
Variable-Gender
Male
Female
Assaults, Thefts and Verbal Abuse Behaviors
2L SD df t p
20.11 8.04
97 0.86 0.39
18.38 7.66
Note. p*<0.05.
Table 8
Independent T-Test for Violent Behaviors Involving Weapons Scale Score
Variable-Gender
Male
Female
Behaviors Involving Weapons
X. SD df t £
3.0 0.0
79 -1.08 0.28
3.13 0.49
Note. p*<0.05.
There was no significant difference in the assaults, thefts and verbal abuse
scores between males (M=20.11, SD= 8.04) and females [M=18.38, SD=7.66,
t(97)=0.858, p=0.39]. The magnitude of the difference in the means was small (eta
squared=.0175). There was also no significant difference in the violent behavior
involving weapons scores between males (M=3.0, SD=0.00) and females [M=3.13,
SD=0.49, t(96)=-1.04, p=0.28].
Results— Number o f years teaching. A one-way between groups analysis of
variance was conducted to explore the relationship between number of years
teaching and teacher perceptions of Violent Behaviors in School— assaults, thefts
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
and verbal abuse and Violent Behaviors in School— involving weapons scale scores.
Subjects were divided into six groups according to the total number of years as a
teacher (“2 or less”, “2-3”, “4-5”, “6-10”, “11-20”, “20 or more”). As shown in
Table 9, there was a statistically significant difference at the p< .05 level in the
assaults, thefts and verbal abuse scale scores for the six groups [F(5, 92)= 2.85,
p= .02]. The actual difference in mean scores between the groups was quite large.
The effect size calculated using eta squared, was 0.13.
Post-hoc comparisons using the Tukey HSD test indicated that the mean
score of teachers with 4-5 years of teaching experience (M=25.67, SD=8.33) were
significantly different from teachers with 6-10 years of teaching experience
(M=15.9, SD=5.78). Teachers with 4-5 years of teaching experience reported
significantly higher assaults, thefts and verbal abuse scale scores than teachers with
6-10 years of teaching experience. The other four groups were not significantly
different.
Similarly, a one-way between groups analysis of variance was conducted to
explore the relationship between number of years teaching and teacher perceptions
of violent behaviors involving weapons. Unlike the Violent Behaviors in School—
assaults, threats and verbal abuse, there was no significant difference at the p< .05
level for the six groups [F(5, 91)= 1.47, p= 0.21]. Although no significant
difference existed between the means, the difference in mean scores was moderate.
The effect size calculated using eta squared, was 0.07.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 9
Comparison o f Teacher Perceptions o f Violent Behaviors in School by Years o f Teaching Experience
2 or less years 2-3 years 4-5 years
N Mean
Std.
Dev. N Mean
Std.
Dev.
Std.
N Mean Dev.
Violent behaviors in school-
assaults and verbal threats
Violent behaviors in school-
weapons________________
20.44
3.0
7.42
0
15.67
3
4.62
0
25.67
3.0
8.24
0
Table 9. (Continued)
6-10 years 11-20 years 20 or more years ANOVA
Std. Std. Std.
N Mean Dev. N Mean Dev. N Mean Dev.
Violent behaviors in school-
assaults and verbal threats 30 15.90 5.78 27 21.11 7.98 23 17.70 8.46 2.79 0.02*
Violent behaviors in school-
weapons______________________ 28 3.0 0_______ 28 3.11 0.31_______23 3.30 0.82________ 1.47 0.21
Note. p*<0.05
oo
SO
90
Results— Number o f years teaching at school site. A one-way between
groups analysis of variance was conducted to explore the relationship between
number of years teaching at that particular school site and teacher perceptions of
Violent Behaviors in School— assaults, thefts and verbal abuse and Violent
Behaviors in School— involving weapons scale scores. Subjects were divided into
six groups according to the total number of years as a teacher (2 or less, 2-3, 4-5, 6-
10, 11-20, 20 or more).
As shown in Table 10, there was no statistically significant difference at the
p< .05 level in Violent Behaviors at School— assaults, thefts and verbal abuse scale
scores for the six groups [F(5, 92)= 1.399, p= .23]. The actual difference in mean
scores between the groups was moderate. The effect size calculated using eta
squared, was 0.07. There was also no significant difference in mean scores at the
p<0.05 level in Violent Behaviors at School— involving weapons scale scores
[F(5, 91)=1.43, p= 0.22]. The effect size for this variable was 0.07, indicating a
moderate effect size.
Results— Level o f school violence. A one-way between groups analysis of
variance was conducted to explore the relationship between the level of school
violence, as determined by suspension rates, and teacher perceptions of Violent
Behaviors in School— assaults, thefts and verbal abuse and Violent Behaviors in
School— involving weapons scale scores. Subjects were divided into three groups
according to the previous year’s suspension rates (low violence, moderate violence,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 10
Comparison of Teacher Perceptions o f Violent Behaviors in School by Years o f Teaching Experience at Particular Site
Violent behaviors in school-
assaults and verbal threats
Violent behaviors in school-
weapons________________
2 or less years
Std.
N Mean Dev.
18 19.83 7.51
18 3.0 0
2-3 years
9 14.44
8 3.0
5.48
0
4-5 years
Std.
N Mean Dev.
Std.
N Mean Dev.
12 20.58
12 3.0
8.24
0
Table 10. (Continued)
6-10 years 11 -20 years
N Mean
Std.
Dev.
Violent behaviors in school-
assaults and verbal threats
Violent behaviors in school-
weapons________________
24
24
16.63
3.04
6.79
0.20
Std.
N Mean Dev.
29 20.28 8.86
29 3.28 0.75
20 or more years
Std”
N Mean Dev.
19.67 5.92
3.17 0.41
ANOVA
1.39 0.23
1.43 0.22
Note. p*<0.05.
SO
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 11
Comparison of Teacher Perceptions o f Violent Behaviors in School by School Violence Levels
Low Moderate High ANOVA
N M SD N M SD N M SD F p
Violent behaviors in school -
assaults, thefts and verbal abuse 32 17.31 6.86 30 21.33 7.81 37 17.76 7.22 2.63 0.08
Violent behaviors in school -
involving weapons 32 3.09 0.53 30 3.17 0.53 36 3.06 0.23 0.52 0.60
Note. p*<0.05.
VO
93
high violence). As shown in Table 11, there was no statistically significant difference
at the p< .05 level in Violent Behaviors in School— assaults, thefts and verbal abuse
scale scores among teachers from the three levels of school violence [F(2, 96)= 2.63,
p= .08]. The actual difference in mean scores between the groups was small. The
effect size calculated using eta squared, was 0.05.
There was also no significant difference at the p<0.05 level in Violent
Behaviors in School— involving weapons among teachers from the three levels of
school violence [F(2,95)=0.52, p=0.60]. The actual difference in mean scores
between the groups was small and calculated using eta squared (q =0.01).
Discussion. Significant mean differences of Violent Behaviors in School—
assaults, thefts and verbal abuse was found between teachers who taught for 4-5
years and 6-10 years. Contrary to the hypothesis of significant differences among
the various groups by gender, years of experience at particular school site, and
level of school violence, there were none.
Existence o f Safe School Elements
Research Question 2: Is there a significant difference in elementary school
teachers’ perceptions of the five safe school elements (school climate,
connectedness, expectations, collaborative community network, and discipline/crisis
plans) between teachers from low violence, moderate violence, and high violence
schools on O’ahu?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
The second research question identified any differences of teacher
perceptions between schools with low, moderate and high level violent incidents and
their perceptions of existing safe school elements. An exploratory and then a
confirmatory factor analysis were conducted for each of the safe school elements in
the questionnaire, except for the Collaborative Community Networking element to
determine reliable scales. The reliability of the scales were analyzed and confirmed
with a Cronbach alpha coefficient above 0.7 (Pallant, 2005).
A one-way between groups analysis of variance was conducted with the
mean scale scores of each of the safe school elements to discover significant
differences of variances between the low, moderate and high violence schools. The
Tukey-Kramer multiple comparison procedure further revealed which variables
significantly differed.
Research Question 2a: Is there a difference in school climate scores for
teachers from low, moderate, and high violence schools?
Hypothesis 2a: As the school violence level increases, less positive school
climate attributes will be reported along with more negative school climate
attributes.
Results— School climate. The 12 items of the School Climate question
(question 11) were subjected to principal components analysis (PCA) using SPSS.
Inspection of the correlation matrix revealed the presence of many coefficients of 0.3
and above. The Kaiser-Meyer-Olkin value was 0.741, exceeding the recommended
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
value of 0.6 and the Bartlett’s Test of Sphericity reached statistical significance
which supports the factorability of the correlation matrix.
Principal components analysis revealed the presence of four components with
eigenvalues exceeding 1, explaining 28.47 percent, 17.96 percent, 9.78 percent, and
8.62 percent of the variance respectively. An inspection of the screeplot revealed a
clear break after the second component. These two components were therefore
retained for further investigation and were supported by the results of a Varimax
rotation. The rotated solution revealed strong loadings on two components and all
variables loading substantially on only one component. The two components solution
explained a total of 46.43 percent of the variance, with Component 1 (Negative
School Climate) contributing 24.42 percent and Component 2 (Positive School
Climate) contributing to 22.01 percent.
Two scales were created after a confirmatory analysis was conducted. The
first included the negative school climate attributes with Cronbach’s alpha reliability
statistic of 0.757 and the second included positive school climate attributes with
Cronbach’s alpha reliability of 0.717, indicating reliable scales.
A one-way between-groups analysis was conducted to explore the difference
in positive school climate scale scores among teachers from low, moderate and high
levels of school violence. There was a statistically significant difference at the p<.05
level in positive school climate scores for the three level of violent schools [F(2,
94)= 4.97, p=0.01]. There was a large difference in mean scores between the groups
with an effect size of 0.10.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
Post-hoc comparisons using the Tukey HSD test indicated that the mean
score for low violence school (M= 18.50, SD=2.49) was significantly different from
moderate violence schools (M= 16.46, SD=2.82). The moderate violence schools
were significantly different from high violence schools (M=18.14, SD=2.76)
although no significant difference existed between the low and high violence
schools. Table 12 summarizes these results.
A one-way between-groups analysis was conducted to explore the difference
of negative school climate scale scores among teachers from low, moderate and high
levels of school violence. There was a statistically significant difference at the p<.05
level in negative school climate scores for the three levels of violent schools
[F(2, 96)=4.025, p=0.02]. There was a moderate difference in mean scores between
the groups with an effect size of 0.08. Post-hoc comparisons using the Tukey HSD
test indicated that the mean score for low violence school (M=9.31, SD=2.11) was
significantly different from high violence schools (M= 10.89, SD=2.50). The
moderate violence schools (M= 10.90, SD= 3.14) did not differ significantly from
either the low or high violence schools. Table 12 summarizes the data.
Discussion. The results of the analysis of variance for the positive school
climate across three violence levels partially support the hypothesis that as the level
of violence increases, the positive school climate score will decrease. The low
violence schools reported significantly higher positive school climate scores than the
moderate violence school as hypothesized. The high violence school, however, had
significantly higher positive school climate scores than the moderate violence school.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced w ith permission o f th e copyright owner. Further reproduction prohibited without permission.
Table 12.
Comparison o f Elementary School Teachers ’ Perceptions o f Safe School Elements Across Violence Levels
Low Violence Moderate Violence High Violence ANOVA
N Mean
Std.
Dev. N Mean
Std.
Dev. N Mean
Std.
Dev. F
P
Positive school climate 32 18.50 2.49 28 16.46 2.82 37 18.14 2.76 4.97 0.01*
Negative school climate 32 9.31 2.12 29 10.90 3.14 38 10.89 2.50 4.03 0.02*
Total Connectedness 32 30.09 2.45 30 28.50 3.37 37 28.76 3.36 2.43 0.09
Total Expectations Score 31 23.61 2.32 26 21.58 2.40 40 22.10 2.60 5.50 0.01*
Total Collaborative Community
Network Score 25 66.52 19.24 24 65.79 12.35 30 70.10 18.39 0.51 0.61
Total Discipline and Crisis Plan Score 30 24.33 3.40 28 20.29 3.31 39 22.33 2.76 12.14 0.00*
Note. p*<0.05.
v o
-j
98
The results of the negative school climate element partially support the
hypothesis that as the school violence level increases, negative school climate scores
will also increase. A significant difference between the mean negative school climate
scale scores of the low violent school and high violence school was found. Although
the mean negative school climate score of the moderate school was similar to that of
the high violence school, no significant difference in means resulted.
Research question 2b. Is there a difference in connectedness scores for
elementary school teachers from low, moderate, and high violence schools?
Hypothesis 2b. As the school violence level increases, teacher connectedness
scale scores will decrease.
Results— Connectedness. The nine items of the Connectedness question
(question 11) were subjected to principal components analysis (PCA) using SPSS.
Inspection of the correlation matrix revealed the presence of many coefficients of 0.3
and above. The Kaiser-Meyer-Olkin value was 0.829, exceeding the recommended
value of 0.6. Principal components analysis revealed the presence of one component
with an eigenvalue exceeding 1 and explaining 46.2 percent of the variance. A
confirmatory factor analysis confirmed the single scale. Therefore, one scale was
created with Cronbach’s alpha reliability statistic of 0.829, indicating a reliable scale.
A one-way between-groups analysis was conducted to explore the difference
in total Connectedness scale scores among teachers from low, moderate and high
levels of school violence. There was no statistically significant difference at the
p<.05 level in connectedness scores for the three level of violent schools
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
[F(2, 96)=2.431, p=0.09]. There was a moderate difference in mean scores between
the groups with an effect size of 0.06. Results are summarized in Table 12.
Discussion. No significant differences exist between the means across school
level violence and therefore, the results reject the hypothesis that as the school
violence level increases, teachers’ connectedness scores will decrease.
Research question 2c. Is there a difference in expectations scores for
elementary school teachers from low, moderate, and high violence schools?
Hypothesis 2c. As the school violence level increases, teacher expectations
scale scores will decrease.
Results- Expectations. The seven items of the expectation question (question
13) were subjected to principal components analysis (PCA) using SPSS. Prior to
performing PCA the suitability of data for factor analysis was assessed. Inspection of
the correlation matrix revealed the presence of many coefficients of 0.3 and above.
The Kaiser-Meyer-Olkin value was 0.808, exceeding the recommended value of 0.6.
Principal components analysis revealed the presence of one component with
an eigenvalue exceeding 1 and explaining 52.23 percent of the variance. Therefore,
one scale was created with Cronbach’s alpha reliability statistic of 0.839, indicating a
reliable scale.
A one-way between-groups analysis was conducted to explore the difference
in expectation scale scores among teachers from low, moderate and high levels of
school violence. There was a statistically significant difference at the p<.05 level in
expectations scores for the three level of violent schools [F(2, 94)=5.501, p=0.01].
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
The difference of mean scores ranged between a moderate and high effect. The effect
size, calculated using eta squared, was 0.10.
Post-hoc comparisons using the Tukey HSD test indicated that the mean
score for low violence school (M=23.61, SD=2.32) was significantly different from
moderate violence schools (M= 21.58, SD=2.40). The low violence school was also
significantly different from the high violence school (M=22.10, SD=2.60). The
moderate violence schools were not significantly different from high violence
schools. Results appear in Table 12.
Discussion. The results of the analysis of variance across school level
violence support the hypothesis that as the level of violence increases, teachers’
reported expectations scores will decrease. As the school violence level increased,
the reported expectations scores decreased between low and moderate violence
schools and between low and high violence schools. This would suggest that teachers
in the participating low violence school reported holding higher expectations for
their students providing them with clear and consistent expectations, multiple
opportunities for success and an expectation that students are motivated to learn.
Further research is needed to explore the differences in expectations between
moderate and high violence schools.
Research question 2d. Is there a difference in collaborative community
networking scores for elementary school teachers from low, moderate, and high
violence schools?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
Hypothesis 2d. As the level of school violence increases, teacher
collaborative community networking scale scores will decrease.
Results— Collaborative community networking. A single scale of 22 items
was created for the collaborative community networking question (question 14). A
Cronbach’s alpha reliability statistic of 0.96 indicates a reliable scale.
A one-way between-groups analysis was conducted to explore the difference
in total collaborative community networking scale scores among teachers from low,
moderate and high levels of school violence. As shown in Table 12, there was no
statistically significant difference at the p<.05 level in collaborative community
networking scores for the three school violence levels [F(2, 976)=0.505, p=0.61].
The difference of mean scores was small. The effect size, calculated using eta
squared, was 0.01. Results are summarized in Table 12.
Discussion. The results of the collaborative community networking element
reject the hypothesis that as the level of school violence increases, teachers’
collaborative community networking score will decrease. No significant difference
in mean collaborative community networking scores between the three levels of
school violence was found.
Research question 2e. Is there a difference in discipline and crisis plan
scores for elementary school teachers from low, moderate, and high violence
schools?
Hypothesis 2e. As the level of school violence increases, teacher scale scores
for the discipline and crisis plans scale will decrease.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
Results- Discipline and crisis plans. The seven items of the discipline and
crisis question (question 15) were subjected to principal components analysis (PCA)
using SPSS. Inspection of the correlation matrix revealed the presence of many
coefficients of 0.3 and above. The Kaiser-Meyer-Olkin value was 0.827, exceeding
the recommended value of 0.6.
Principal components analysis revealed the presence of one component with
an eigenvalue exceeding 1 and explaining 67.63 percent of the variance. Therefore,
one scale was created with Cronbach’s alpha reliability statistic of 0.921, indicating a
reliable scale.
A one-way between-groups analysis was conducted to explore the difference
in discipline and crisis plans scale scores among teachers from low, moderate and
high levels of school violence. There was a statistically significant difference at the
p<.05 level in discipline and crisis plans scores for the three level of violent schools
[F(2, 94)=12.137, p=0.00]. There was a large difference in mean scores between the
groups. The effect size, calculated using eta squared, was 0.21. Results are
summarized in Table 12.
Post-hoc comparisons using the Tukey HSD test indicated that the mean
score for low violence school (M=24.33, SD=3.40) was significantly different from
moderate violence schools (M= 20.29, SD=3.31). The low violence school was also
significantly different from the high violence school (M=22.33, SD=2.76).
Additionally, the moderate violence schools were not significantly different from the
high violence school in mean discipline and crisis plans scores.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
Discussion. The results of the analysis of variance across school level
violence support the hypothesis that as the level of violence decreases, teacher scores
for the discipline and crisis plans scale increases. As the school violence level
increased, the reported discipline and crisis scale scores decreased between low and
moderate violence schools and between low and high violence schools. This would
suggest that teachers in the participating low violence school reported higher
amounts of fair treatment of students by teachers and school leadership, and the
existence of clear discipline plans made known to both teachers and students. Further
research is needed to explore the differences in discipline and crisis scale scores
between moderate and high violence schools.
Responsibility for Safe Schools
Research question 3. To what degree do elementary school teachers perceive
the stakeholders (teacher, administrator, parent, and student) responsible for
contributing to safe schools?
Results. Table 13 summarizes the results of teacher perceptions of how much
responsibility teachers placed on each of the seven stakeholders in education. More
teachers placed the highest responsibility on administrators (68.8%— very much),
followed by teachers (52%— very much). Additionally, 46 percent of teachers
responded that students should take “very much” of the responsibility for ensuring
safe schools.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
Table 13
Percentage o f Teachers ’ Perceived Responsibility for Ensuring Safe Schools
None A little Some Quite a bit Very much
Administrators 0 0 4.2 27.1 68.8
Teachers 0 0 7.1 40.8 52.0
Parents 1.0 6.1 25.5 34.7 32.7
Students 0 4.1 16.3 33.7 45.9
Community 3.1 6.1 31.6 31.6 27.6
Resource Teachers
and Complex Area
Supervisors
10.4 9.4 31.3 26.0 22.9
DOE State Level
Supervisors
13.4 9.3 22.7 26.8 27.8
Discussion. Teachers placed less responsibility on stakeholders who did not
have opportunities for everyday interactions with the students such as the
community, resource teachers and complex area supervisors and the Department of
Education level supervisors/personnel. These stakeholders also reported the highest
percentages in the “none” and “a little” categories.
Research question 3a: How does perceived responsibility for providing safe
schools vary amongst schools with low, moderate, and high reported violent
incidents?
Hypothesis 3a: As school violence levels increase, teachers report lower
perceived responsibility for each of the stakeholders (administrator, teachers,
parents, students, community, resource teachers and complex area supervisors, and
DOE state level supervisors/personnel).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
Results. Question 25 on the survey instrument was analyzed and responses
were separated by school violence levels. Descriptive statistics were analyzed and
reported as a crosstabulations table showing the teacher responses from each of the
three levels of school violence. The results were compared across all three school
violence levels.
A similar amount of teachers from all three levels indicated that
administrators were very much responsible for ensuring safe schools, 68.8 percent,
67.9 percent and 69.4 percent respectively. More teachers from the low violence
school indicated that teachers were very much responsible for ensuring safe schools
(59.4 %) followed by teachers from the moderate violence schools (50.0 %) and high
violence schools (47.2 %). Data presented in Table 14 show a similar relationship for
responsibility placed on parents. Teachers from the low violence school perceived
parents as very much responsible for ensuring safe schools (46.9%), 26.7 percent
from the moderate violence schools and 25.0 percent from the high violence school.
More teachers from the low violence school perceived students as very
responsible for ensuring safe schools (56.3%) compared to moderate violence
schools (40.0%) and the high violence school (41.7%). Also, more teachers from the
low violence school perceived the community and Department of Education level
supervisors/personnel as very much responsible for school safety than teachers from
moderate and high violence schools.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
Table 14
Percentage o f Teachers ’ Perceived Responsibility for Ensuring Safe Schools by
School Violence Level
None A little Some Quite a bit Very much
Low Violence School:
Administrators 0 0 0 31.3 68.8
Teachers 0 0 6.3 34.4 59.4
Parents 0 0 21.9 31.3 46.9
Students 0 0 12.5 31.3 56.3
Community 0 6.3 21.9 34.4 37.5
Resource Teachers 12.9 6.5 32.3 22.6 25.8
and Complex Area
Supervisors
DOE State Level 12.9 9.7 19.4 22.6 35.5
Supervisors
Moderate Violence
School:
Administrators 0 0 14.3 17.9 67.9
Teachers 0 0 3.3 46.7 50.0
Parents 3.3 10.0 26.7 33.3 26.7
Students 0 3.3 13.3 43.3 40.0
Community 10.0 10.0 36.7 26.7 16.7
Resource Teachers 17.2 17.2 17.2 34.5 13.8
and Complex Area
Supervisors
DOE State Level 23.3 13.3 13.3 33.3 16.7
Supervisors
High Violence School:
Administrators 0 0 0 30.6 69.4
Teachers 0 0 11.1 41.7 47.2
Parents 0 8.3 27.8 38.9 25.0
Students 0 8.3 22.2 27.8 41.7
Community 0 2.8 36.1 33.3 27.8
Resource Teachers 2.8 5.6 41.7 22.2 27.8
and Complex Area
Supervisors
DOE State Level 5.6 5.6 33.3 35.0 30.6
Supervisors
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
Discussion. The results in Table 14 support the hypothesis that as school
violence levels increase, teachers report lower perceived responsibility for teachers
and parents. In addition, more teachers from the low violence school perceived
students, community and DOE state level supervisors/personnel as “very much”
responsible for school safety than teachers from moderate and high violence schools.
All three school violence levels placed most responsibility on stakeholders
who are involved with students on an everyday basis and who have daily decision
making capabilities such as administrators, teacher, and students.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
108
CHAPTER 5
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
This chapter includes an introduction, a review of the purpose of the study, a
summary of the findings, and conclusions followed by recommendations.
Introduction
Despite media portrayals of fatal incidents in American schools, violence is
in fact on a downward trend (Devoe et al., 2004). Although Hawai’i has fortunately
been without fatal school shootings, the trend of violence in Hawai’i public schools
has followed that of its mainland counterpart. Violent incidents in Hawai’i schools
have steadily declined since 1995-1996 until 2003-2004 when discrepancies were
found in the statewide reporting system (Department of Education State of Hawai’i,
2004). Such discrepancies in state reporting along with the ambiguity of federal
reporting requirements have made it difficult to portray the most accurate picture of
school safety in Hawai’i.
The literature produced a limited amount of research on elementary school
personnel perceptions of violence, however, none was specific to Hawai’i.
Therefore, the need existed for a focus on elementary school teacher perceptions to
determine the pervasiveness of violence in Hawai’i public education. This is
imperative in order to work towards developing a more preventative approach to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
decreasing violence in the schools. In addition, there has not been a study that
investigated the presence of the five safe school elements as proposed in this study’s
comprehensive conceptual model. The results of this study will equip policymakers
and administrators with a plan to implement the appropriate assistance in helping all
stakeholders in education understand their roles in preventing school violence.
The Purpose of the Study
The purpose of this study was to examine the perceptions of elementary
school teachers on school violence in O’ahu, Hawai’i. This study also investigated
the perceived existence of the five elements of safe schools across schools with low,
moderate and high violence levels as determined by each school’s suspension rates.
Lastly, this study examined the perceived responsibility for providing safe schools
among all stakeholders in education. The data were analyzed according to the
research questions being investigated.
Summary of Findings
There were several focuses of this study—overall perceptions of school
violence, presence of safe school elements, and perceived responsibility of school
safety. The following will summarize the findings of the data analysis by research
question.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
Perceptions o f School Violence
The results of the observed risk behaviors and violence behaviors in school in
the teacher questionnaire reported similar results. High reports of bullying,
vandalism, stealing and perceptions that students get into fights were similar to the
results of violent behaviors in school. Mocking and insulting and threats to hurt
another are forms of bullying and thefts are equivalent to stealing.
Violent Behaviors—gender, years o f teaching experience, years at school site,
school violence level.
As hypothesized, less experienced teachers, reported a greater frequency of
violent behaviors in school— assaults, thefts and verbal abuse. Teachers with 4-5
years of teaching experience reported significantly higher assaults, thefts and
verbal abuse scale scores than teachers with 6-10 years of teaching experience.
Surprisingly, there were no significance found between gender, years at that
particular school site and school violence level. There were also no significant
differences in mean scale scores of violence behaviors in school— involving
weapons for either of the four variables.
Presence o f Safe School Elements
Significant differences across the three school violence levels were found for
the following safe school elements—school climate, expectations, discipline and
crisis plans. Surprisingly no significant difference between mean scale scores across
the three violence levels were found for connectedness, and collaborative community
networking.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I l l
Significant differences were found for the two reliable scales developed for
the school climate element— positive school climate attributes and negative school
climate attributes. The low violence schools reported significantly higher positive
school climate scores than the moderate violence school as hypothesized. The high
violence school, however, had significantly higher positive school climate scores
than the moderate violence school.
A significant difference between the mean negative school climate scale
scores of the low violent school and high violence school was found. No significant
difference in means resulted between the moderate and high violence schools.
As the school violence level increased, the reported expectations scores
decreased between low and moderate violence schools and between low and high
violence schools. Teachers from the low violence school reported higher
expectations for their students than teachers from moderate violence schools and
teachers from the high violence school. No significance was found between the mean
expectations scale scores of the moderate and high violence schools.
The results also revealed important significant findings for reported
perceptions of the presence of discipline and crisis plans. As the school violence
level increased, the perceived presence of discipline and crisis plans decreased
between low and moderate violence schools and between low and high violence
schools. Therefore, teachers in the participating low violence school reported higher
amounts of fair treatment of students by teachers and school leadership, and the
existence of clear discipline plans made known to both teachers and students than
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
112
their counterparts in both the moderate and high violence schools. Further research is
needed to explore the differences in discipline and crisis scale scores between
moderate and high violence schools.
Responsibility fo r Safe Schools
Descriptive statistics reported that teachers overall viewed the following
stakeholders in education as “very much” responsible for ensuring safe schools in the
following order: administrators, teachers, students and parents. Teachers placed less
responsibility on stakeholders who did not have opportunities for everyday
interactions with the students such as the community, resource teachers and complex
area supervisors and the Department of Education level supervisors/personnel.
The second question pertaining to responsibility for safe schools investigated
a comparison of responses across all three levels of violence. Interestingly, as the
school violence levels increased, teachers reported lower perceived responsibility for
the following stakeholders— teachers and parents. In addition, more teachers from
the low violence school perceived students, community and DOE state level
supervisors/personnel as “very much” responsible for school safety than teachers
from moderate and high violence schools.
Similarly to the overall perceptions of teachers in regards to responsibility, all
three school violence levels placed most responsibility on stakeholders who are
involved with students on a daily basis and who have daily decision making
capabilities such as administrators, teachers, and students.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
113
Conclusions
Most elementary teachers perceived their schools as safe where violence is a
“very small problem” or “not a problem at all.” This supports what has been reported
in the limited literature on violence in the elementary schools. For example, studies
conducted by Devoe et al. (2004) and Crosse et al. (2000) reported an inverse
relationship between school level and violent crime involving student victimizations.
The results also support previous research that higher occurrences of bullying are
presented in the elementary school level (Addington et al., 2002; Devoe et al., 2004;
Furlong, & Morrison, 2000). Teachers reported more incidents of bullying than any
other violent behavior and these results will be useful in determining the next steps
in ensuring safer schools.
With regards to the presence of the five safe school elements, there were
surprisingly no significance found across the levels of violence for two of the
prevention elements— connectedness, and collaborative community networking. This
contradicts what has been found in the majority of the literature. For example, a
study by Wilson (2004) found that students with high connectedness are less likely to
experience high levels of victimizations and aggression. This is coupled with results
of Astor et al. (1999) that teachers who are connected to their students are more
likely to take responsibility for school campuses. It is even more surprising that no
significance was found for connectedness because school climate encompasses the
element of connectedness.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
114
The results for the connectedness element of safe schools may be due to
several possibilities. First, the mean connectedness scale scores of schools from all
three violence levels ranged from 28.50 to 30.09 out of a possible score of 36.
Although there were no significant differences between the school violence levels,
the results are nonetheless important. There were no significant differences possibly
because all schools reported relatively high scale scores suggesting that
connectedness is present at these sample elementary schools. It would seem
reasonable for elementary teachers to have high connectedness scores considering
they remain with their classroom students throughout the day. Second, the
connectedness results may be contributed to the findings of Wilson (2004) that both
school climate and school connectedness act almost independently in predicting
aggression. Therefore, the significance found between the means of the school
climate scale scores may account for some characteristics also found in the
connectedness element.
The three significant elements of safe schools presented interesting results as
well. Significant differences were found between the low and moderate violence
schools in regards to positive school climate attributes. As hypothesized, there was
an inverse relationship. On the other hand, the significant difference in the positive
school climate mean scale scores between the moderate and high violence schools
represented a positive relationship. This relationship could be explained by several
possibilities. First, between-school differences for the three moderate schools
examined may have neutralized the effect that a single school may have had. Second,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
115
the use of suspension rates as the indicator of school violence levels may not be a
valid measure. Third, differences of outside factors such as social-economic status,
school size and the location of the school as supported by the literature may have
affected the results. Last, a review of the sample school demographics provided
valuable insight. There were more teachers in the low violence school with advanced
degrees, followed by teachers in the moderate schools and then by teachers in the
high violence school. In addition, there were more teachers in the low violence
school who were fully licensed with an almost equal amount to fully licensed
teachers in the moderate and high violence schools. These factors may have
influenced the results in some way.
Similar results were presented for the expectations and discipline and crisis
plans elements where significance were found between the low and moderate
violence schools and the between the low and high violence schools. In all cases,
there was an inverse relationship between scale scores and level of school violence.
Surprisingly, there was no significance found between the moderate and high
violence schools.
The non-significance found between the moderate and high violence schools
may be attributed to several possibilities. First, the between-school differences for
the three moderate schools examined may have neutralized the effect that a single
school may have had. Second, the use of suspension rates as the indicator of school
violence levels may not be a valid measure. Third, the number of schools utilized to
represent each school violence level may not have been large enough to represent a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
116
true representation of the population being examined. Fourth, the sample
demographics provided insight into outside factors that may have affected the results
such as the number of fully licensed teachers, the amount of teachers with advanced
degrees, and the percentage of students receiving free or reduced lunch. Last,
differences between the moderate and high violence schools may be too similar to
measure significant differences.
In regards to the responsibility for ensuring safe schools, the results raised
further questions that can be addressed in future research. Surprisingly, fewer than
half of the teachers in the moderate and high violence schools thought they bore
primary responsibility for ensuring safe schools, 40 percent and 41.1 percent
respectively in comparison to 56.3 percent of teachers from the low violence school.
Perhaps the lower perceived responsibility of teachers from the moderate and high
violence schools have an affect on how these teachers respond to violent behaviors.
Behre et al. (2001) explored elementary teacher perceptions of their reasoning to
intervene in a physical fight at school and found that location along with a sense of
personal responsibility does matter. Therefore, this question needs to be further
explored to identify the root to why and locations where teachers feel most
responsible for student safety.
In addition, less than half of the teachers from the moderate and high
violence schools placed primary responsibility for safety on students themselves,
40.0 percent and 41.7 percent respectively in comparison to 56.3 percent of teachers
from the low violence school. The question raised here is, “Does having high
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
117
expectations for students related to holding students accountable for ensuring safe
schools?” The perception of not seeing these students as responsible agents may be
the problem and may possibly affect the level of violence in the schools.
Recommendations
To the extent that these findings are generalizable to other public elementary
schools on O’ahu, there are implications for the individual school administrators and
for the Hawai’i Department of Education. The data, however, must be interpreted
with caution as it consists of a small sample size and is solely based on suspension
rates to determine school violence levels.
Recommendations fo r Individual School Site
Based on the findings and conclusions of this study, the following
recommendations are made for the individual sample schools.
• Increase teacher awareness of school violence at individual school sites.
School administrators should share the significant findings with their
faculty, staff and community.
• Further investigate the high reported rates of bullying. Identify specific
incidents that commonly occur at the individual schools. Address the
behaviors appropriately.
• Utilize the significant findings from the teacher survey, SSVSSE, to
develop or enhance an existing school-wide violence prevention program
that includes five of the safe school elements.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
118
• Review and enhance existing discipline and crisis plans at each individual
school site. Increase school staff awareness of the significance of this
element on school violence.
Recommendations for District Level Action
Based on the findings and conclusions of this study, the following
recommendations are suggested for district level action.
• Compare data of teacher perceptions to anchors that include student
responses and documentation of student behavior to assess a more
accurate portrayal of the school violence problem district-wide.
• Assess the current teacher training programs on the topic of school safety
to get an indication of possible areas to strengthen.
• Utilize the results to guide teacher training programs to include the five
safe school elements.
• Investigate the reported difference between teachers with varying years of
experience (6-10 years and 4-5 years). Address the reasons for these
differences with appropriate professional development.
• Develop and implement training on responsibility for school safety and
the role of the stakeholders in education for providing safe schools.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
119
Recommendations fo r Future Research
Based on the findings and conclusions of this study, the following
recommendations are made for future research.
• Include all elementary schools in the state in the sampling to assess
teacher perceptions statewide. Include and compare findings to student
perceptions and school documentations of student behaviors for the most
reliable data.
• Determine the influence of other demographic variables on the teacher
perceptions of violent behaviors. For example, differences between the
five sample schools were found in the following variables, social
economic status, teachers with advanced degrees and teachers the
percentage of fully licensed teachers. Include ethnicity as a variable to
investigate the extent that cultural norms influence results.
• Compare the data with data collected in two parallel studies conducted in
the middle and high schools to investigate the differences between the
three levels of education. Collapse data across all three studies to develop
a comprehensive and seamless approach to preventing school violence
state-wide in all levels of public education.
• Replicate this study at a national level to determine overall teacher
perceptions of violence in schools, the existence of the elements of safe
schools and responsibility. Replace the use of suspension rates as an
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
indicator of violence level with student self-reports and school
documentation of student behaviors.
• Further investigate the relationship between personal responsibility and
the issue of accountability to ensure safe schools. Increase sample size to
determine significant differences between the violence school levels.
Although research shows the decline of school violence in the public schools
nation-wide, violence continues to pervade the environment set for educating our
children. This study presented a comprehensive conceptual model of safe schools to
address violence in the elementary schools. While results show significant results for
three of the five safe school elements, the data also revealed the importance of the
other two elements in developing a school-wide violence prevention program.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
121
References
Addington, L.A, Rudder, S. A., & Chandler, K.A. (2002). Are America’ s schools
safe? Students speak out: 1999 School crime supplement statistical analysis
report (NCES 2002-33 l).Washington, DC: U.S. Department of Education,
Office of Educational Research and Improvement.
The American Heritage® dictionary o f the English language (4th Ed.).
(2000). Houghton Mifflin Company. Retrieved March 21, 2005 from
www.dictionary.com
Astor, R.A. (1995). School violence: A blueprint for elementary school
interventions. Social Work in Education, 17(2), 101-116. Retrieved on
October 12, 2004, from EBSCO database.
Astor, R.A., Behre, W.J., Fravil, K.A., & Wallace, J.M. (1997). Perceptions of
school violence as a problem and reports of violent events: A national survey
of school social workers. Social Works, 42(1), 55-68.
Astor, R.A., Benbenishty, R., Pitner, R.O., & Meyer, H.A. (2003). Bullying and peer
victimization in schools. In P.A. Meares and M.W. Fraser (Eds.),
Intervention with Children & Adolescents: An Interdisciplinary
Perspective, (pp. 471-448). Allyn & Bacon.
Astor, R.A., Meyer, H.A., & Behre, W.J. (1999). Unowned places and times: Maps
and interviews about violence in high schools. American Educational
Research Journal, 56(1), 3-42.
Astor, R.A, Meyer, H., Benbenshity, R., Marachi, R., & Rosemond, M. (2005).
School safety interventions: Best practices and programs. Children and
Schools, 27(1), 17-32.
Batsche, G., & Knoff, H. (1994). Bullies and their victims: Understanding a
pervasive problem in schools. School Psychology Review, 23(2), 165-175.
Retrieved on March 21, 2005 from EBSCO database.
Benard, B. (1991). Fostering resiliency in kids: Protective factors in the family,
school, and community. (1991). Retrieved on March 23, 2005 from
http ://www.nwrac. org/pub/library/f/ f_foster.
Benard, B. (1993). Fostering resiliency in kids. Educational Leadership, 51(3),
44-48.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
122
Benard, B. (1997). Turning it around for all youth: From risk to resilience (Report
No. EDO-UD-97-7). Washington, DC.: Office of Educational Research and
Improvement. (ERIC Document Reproduction Service No. ED412309)
Benbenishty, R., & Astor, R.A. (2005). School Violence in Context: Culture,
Neighborhood, Family, School, and Gender. New York, NY: Oxford
University Press.
Bennett-Johnson, E. (2004). The Root of school violence: Causes and
recommendations for a plan of action. College Student Journal, 35(2).
Retrieved October 12,2004 from EBSCO Database.
Board of Education. (1995). Student code o f conduct Statute 4010. Retrieved August
4, 2004, from http://lilinote.kl2.hi.us/STATE/BOE/POLl.NSF/
85255a0a0010ae82852555340060479d/6aa6673a5e082caa0a2566a300679c6
7?OpenDocument
Bonny, A.E., Britto, M.T., Klostermann, B.K., Homung, R.W., & Slap, G.B. (2000).
School disconnectedness: Identifying adolescents at risk. Pediatrics, 106(5),
1017-1021.
Borja, R.R. (2005, March 22). Details of Minn, school shooting emerge. Education
Week. Retrieved on March 23,2005, from http://www.edweek.org
Borja, R.R., & Cavanagh (2005, March 24). School shootings stun reservation.
Education Week. Retrieved on April 2, 2005, from http://www.edweek.org
Brannon, J. (2005, February 15). School violence difficult to assess. The Honolulu
Advertiser. Retrieved March 18, 2005, from
www.thehonoluluadvertiser.com/article/2005/Feb/15/ln/ln0 lp.html
Boulter, L. (2004). Family-School Connection and School Violence Prevention. The
Negro Educational Review, 55(1), 27-40.
Chesney-Lind, M., & Nakano, J. (2003). How safe are Hawai ’ i ’ s Schools? An
overview o f challenges in Hawai ’ i ’ s Public Schools. Retrieved on December
10, 2004 from http://www.chesneylind.com/reports/ssm_voll.pdf
Christie, C.A., Jolivette, K., & Nelson, C.M. (2000). Youth Aggression and violence:
Risk, resilience, and prevention (Report No. EDO-EC-OO-11). Washington,
DC.: Office of Educational Research and Improvement. (ERIC Document
Reproduction Service No. ED449632)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
123
Collier, K., Kelling, I.K., Siaosi, L. A. (2005). The Survey o f School Violence and
Safe School Elements (SSVSSE).
Colvin, G., & Kameenui, E.J. (1993). Reconceptualizing behavior management and
school-wide discipline in general education. Education & Treatment o f
Children, 76(4), 361-381.
Comer, J.P., & Haynes, N.M. (1991). Parent Involvement in schools: An Ecological
approach. The Elementary School Journal, 9/(3), 271-297.
Crosse, S., Burr, M., Cantor, Burr, M., Hagen, C.A., Hantman, I., et al. (2002). Wide
scope, questionable quality: three reports from the study o f school violence
and prevention. (U.S.DOE Contract Number EA96055001). Washington,
DC: U.S. Department of Education, Planning and Evaluation Service.
Department of Education State of Hawaii. (2004). Introduction, school safety.
Retrieved on August 2, 2004, from
http://doe.kl2.hi.us/about/intro_safety.htm
Department of Education State of Hawai’i. (2005). Mandatory Expulsion Policy for
Possession o f a Firearm School Year 2003-2004. Legislative Report.
Retrieved on March 21, 2005, from
http://doe.kl2.hi.us/reports/tolegislature/MandatoryExpulFirearmPossession.
pdf
Department of Education State of Hawai’i. (2004). The Superintendent’ s Fourteenth
Annual Report on School Performance and Improvement in Hawai ’ i 2003.
Office of the Superintendent/Planning and Evaluation Office Available
online at http://arch.kl2.hi.us/pdf/report/2003/SuptRept2003.pdf.
Department of Education. (2002). Wide scope, questionable quality executive
summary.(Contract No. EA96055001). Rockville, MD: Author.
Department of Education Office of Elementary and Secondary Education. (2002,
July 2). Notice authorizing schoolwide programs to consolidate federal
education funds and exempting them from complying with statutory or
regulatory provisions o f those program s. Federal Register, 69 (127).
Devoe, J.F., Peter, K., Kaufman, P., Miller, A., Noonan, M., Snyder, T.D. et al.
(2004). Indicators o f school crime and safety: 2004 (NCES Publication No.
2005-002/NCJ Publication No. 205290). U.S. Department of Education and
Justice. Washington, DC: U.S. Government Printing Office. Available online
at http://www.ojp.usdoj.gov/bjs/pub/pdf/iscs04.pdf
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
Dwyer, K., & Osher, D. (2000). Safeguarding Our Children: An Action Guide.
Washington, D.C.: U.S. Departments of Education and Justice, American
Institutes for Research. Available at
http://www.ed.gov/offices/OSERS/OSEP/ActionGuide
Dwyer, K, Osher, D., & Warger, C. (1998). Early Warning, Timely Response.
Washington, DC: U.S. Department of Education. Available online at
http://www.ed.gov/offices/OSERS/OSEP/earlywm.html
Ed.gov U.S. Department of Education. The Facts about school safety. Retrieved
July 30, 2004, from http://www.ed.gov/nclb/admins/lead/safety/
Eisenberg, M.E., Neumark-Sztainer., & Perry, C.L. (2003). Peer harassment, school
connectedness, and academic achievement. Journal o f School Health, 73(8),
311-316. Retrieved on March 23, 2005, from EBSCO database.
Elementary and Secondary Education Act, Title IV- 21st Century Schools. Retrieved
August 1, 2004, from http://www.ed.gov/policy/elsec/leg/esea02/pg51.html
Essoyan, S. (2003, July 6). Hawai’i Public Schools: Safer to enter? Honolulu Star
Bulletin. Retrieved February 6,2005, from
http://starbulletin.com/2003/07/06/news/story2.html
Fein, R.A., Vossekuil, B., Pollack, W.S., Borum, R., Modzeleski, W., & Reddy, M.
(2002). Threat assessment in schools: A Guide to managing threatening
situations and to creating safe school climates. Washington, D.C.: United
States Secret Service and United States Department of Education.
Fujimori, L., & Martin, D. (2005, February 8). Nanakuli students continue brawling
[Electronic Version], The Honolulu Star Bulletin.
Furlong, M., & Morrison, G. (2000). The School in school violence: Definition
and facts. Journal o f Emotional & Behavioral Disorders, 5(2). Retrieved on
March 16, 2005 from EBSCO database.
Gao, G., & Perrone, P. (2004) Crime in Hawai ’ i 2003: A Review o f uniform crime
reports. Honolulu, HI: Crime Prevention & Justice Division, Research &
Statistics Branch.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
125
Grunbaum, J.A., Kann, L., Williams, B., Ross, J.G., Lowry, R. et al. (2002). Youth
risk behavior surveillance- United States, 2001, Retrieved on March 22,
2005, from http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5104al.htm
Hawaii Department of Education (2004) Hawaii public schools, 2003-2004 Number
and type o f schools. Retrieved August 1, 2004, from
http://doe.kl2.hi.us/reports/schooltypes.htm
Hawai’i Department of Education. (2005). Hawai’i Department o f Education Trend
Report: Educational Fiscal Accountability. Available online at
http://arch.kl2.hi.us
Hawaii Department of Education Office of Curriculum, Instruction and Student
Support. (2003). Safe and drug free schools and communities. Retrieved
August 3, 2004, from http://ociss.kl2.hi.us/ociss/page.jsp?page=program
detail&department=isb&docid=8 5
Hawai’i Educational Policy Center. (2003). Just the facts... A Citizen’ s primer on
Hawai’ i K-12 Public Education. Hawai’i Public Education Center, College of
Education, University of Hawai’i.
Hawkins, J.D., Herrenkohl, T.I., Farrington, D.P., Brewer, D., Catalano, R.F.,
Harachi, T.W. et al. (2000). Predictors of youth violence. Juvenile Justice
Bulletin. Washington, DC: U.S. Department of Justice, Office of Juvenile
Justice and Delinquency Programs. Retrieved on May 25, 2005 from
www.ncjrs.org/pdffilesl/ojjdp/179065.pdf
Haynes, N.M. (1996). Creating safe and caring school communities: Comer school
development program schools. The Journal o f Negro Education, 65(3),
308-314.
Haynes, N.M., & Comer, J.P. (1996). Integrating schools, families, and communities
through successful school reform: The School Development Program. School
Psychology Review, 25(4), 501-507. Retrieved on March 23, 2005, from
EBSCO database.
Haynes, N.M., Hamilton-Lee, M., & Comer, J.P. (2001). Differences in self-concept
among high, average, and low achieving high school sophomores. The
Journal o f Social Psychology, 128(2), 259-264.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
126
Henson, L., Massey, O., & Armstrong, K. (2002). Student perceptions o f school
safety: Results o f the 2001 Omnibus Survey. Tampa, FL: The Louis de la
Parte Florida Mental Health Institute, University of South Florida. FMHI
Publication #207-4.
Hunt, M.H., Meyers, J., Davies, G., Meyers, B., Grogg, K.R., & Neel, J. (2002). A
Comprehensive needs assessment to facilitate prevention of school drop out
and violence. Psychology in the Schools, 39(4), 399-415. Retrieved on
March 23, 2005, from EBSCO database.
Johnson, K. (2005, March 25). Survivors of high school shooting rampage left with
injuries and questions. The New York Times. Retrieved on April 2, 2005,
from http://www.nytimes.com
Kadel, S., Watkins, J., Follman, J., & Hammond, C. (1999). Reducing school
violence: Building a framework for school safety (3rd Ed.) (ED 435 111).
Tallahassee, FL.: Florida State Dept, of Education.
Kellam, S.G., Prinz, R., & Sheley, J.F. (2000). Preventing school violence: Plenary
papers o f the 1999 conference o f criminal justice research and evaluation-
enhancing policy and practice through research, volume 2 (NCJ 180972).
Washington, DC.: U.S. Department if Justice, Office of Justice Programs.
Kinergy, P.M., & Coggershall, M.B. (2001). School-based surveillance o f violence,
injury, and disciplinary actions. Hamilton Fish Institute. Available online at
http://hamfish.org.
Larson, Eric. 2003. Violence in U.S. public schools: A Summary o f Findings. (Report
No. EDO-UD-03-13). Washington, DC: Institute of Education Sciences.
(ERIC Document Reproduction Service No. ED482921)
Leinhardt, A.M., & Willert, H.J. (2002). Involving stakeholders in resolving school
violence. National Association o f Secondary School Principals, 56(631),
p. 32, Retrieved on March 20, 2005 from FirstSearch database.
Leone, D. (2005, February, 5). Campus fighting strikes 3 schools. Honolulu Star-
Bulletin News. Retrieved on February 13, 2005, from
http://starbulletin.com/2005/02/05/news/storyl.html
Levine, D.M., Krehbiel, T.C., & Berenson, M.L. (2003). Business Statistics: A First
course (3rd ed.). New Jersey: Prentice Hall.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
127
Lohrmann, S., & Talerico, J. (2004). Anchor the boat: A classwide intervention to
reduce problem behavior. Journal o f Positive Behavior Interventions, 6(2),
113-120.
Martin, D. (2005, February 1). Threat of violence paralyzes Radford. Honolulu Star-
Bulletin News. Retrieved on February 13, 2005, from
http://starbulletin.com/2005/02/01/news/story3.html
Merriam-Webster. (1996). Merriam- Webster's dictionary [Electronic version].
Springfield, MA: Merriam-Webster. Retrieved July 30, 2004, from
www.dictionary.com
Miller, A. (2003). Violence in U.Spublic schools: 2000 school survey on crime and
safety statistical analysis report. (Report No. NCES-2004-314). Washington,
DC: National Center for Education Statistics (ED). ERIC Document
Reproduction Service No. ED482414)
National Education Goals Panel. (2000). Goal 7: Safe, disciplined, and alcohol- and
drug-free schools. Washington, DC: Author.
No Child Left Behind Act of 2001. (2002). Ed.Gov. Retrieved August 1,2004, from
http ://www.ed. gov/policy/ elsec/leg/ esea02/index.html
O’Donnell, J., Hawkins, J.D., & Abbott, R.D. (1995). Predicting serious delinquency
and substance use among aggressive boys. Journal o f Consulting and
Clinical Psychology, 63(4), 529-537.
Office of Educational Research and Improvement. (2002, Nov.) Indicators o f School
Crime and Safety, 2002.
Office of Curriculum, Instruction and Student Support/Student Support Services
Branch Department of Education. (2002) Chapter 19: Student Misconduct,
Discipline, School Searches and Seizures, Reporting Offenses, Police
Interviews and Arrests, and Restitution for Vandalism and Negligence.
[Brochure].
Outline o f programs and selected changes in the No Child Left behind Act o f2001.
(2002). Retrieved August 1, 2004, from
http://www.ed.gOv/nclb/overview/intro/progsum/sum_pg9.html#drug. 1-121.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
128
Olweus, D. (1991). Bully/victim problems among schoolchildren: Basic facts and
effects of a school-based intervention program. In D.J. Pepler & K.H. Rubin
(Eds.), The development and treatment o f childhood aggression (pp. 411-
448). Hildside, NJ: Lawrence Erlbaum.
Pallant, J. (2005). SPSS Survival manual (2n d ed.). Chicago: Open University Press.
Peterson, R.L., & Skiba, R. (2001). Creating school climates that prevent school
violence. The Social Studies, 92(4), 167-175.
Pollack, I., & Sundermann, C. (2001). Creating safe schools: A comprehensive
approach. Juvenile Justice, 8 (1),73-20. Retrieved on January 12, 2004 from
http ://www.ncjrs .org/pdffiles 1 / ojj dp/18815 8.pdf
Potter, B.B., Beaulieu-Cooke, N.B., Davis, M., Hasbrouck, L.M., Rosenfeld, A.H., &
Tuma, F. (2001). Youth Violence: A Report o f the Surgeon General.
Department of Health and Human Services. Available online at
http://www.surgeongeneral.gov/library/youthviolence/default.htm
Proximity. (Sept. 13, 2003) Hawaii Ranks 10th Among Largest U.S. 25 School
Districts. Retrieved on March 22, 2005, from
http://www.proximityone.com/sdsttop25.htm
Quality counts 2004 count me in: Special education in an era of standards, School
Climate. (2004). [Electronic Version] Education Week, 25(17), 115
Reno, J., Fisher, R.C., Robinson, L., & Bilchick, S. (1999). Promising strategies to
reduce gun violence. U.S. Department of Education and Justice. Office of
Juvenile Justice and Delinquency Prevention. Washington, DC: U.S.
Government Printing Office. Available online at
http://www.ojjdp.ncjrs.org/pubs/gun_violence/173950.pdf
Resnick, M.D., Bearman, P.S., Blum, R.W., Bauman, K.E., Harris, K.M., Jones, J. et
al. (1997). Protecting adolescents from harm: Findings from the national
longitudinal study on adolescent health. Journal o f the American Medical
Association, 278, 823-832.
Resnick, M.D., Harris, L.J., & Blum, R.W. (1993). The impact if caring and
connectedness on adolescent health and well-being. Family Youth and Family
Consortium Website. Retrieved on March 24, 2005, from
http://www.cyfc.umn.edu/schoolage.resources/caring.html
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
129
Rosenblatt, J., & Furlong, M. (1997). Assessing the reliability and validity of
student self-reports of campus violence. Journal o f Youth & Adolescence,
26(2). 187-202. Retrieved on March 21, 2005 from FirstSearch database.
Safe & Drug-Free Schools and Communities Program. (2002). Retrieved
Aug 3,2004, from http://t3.kl2.hi.us/t301-02/lmurakami/drug/drug/
Schwartz, W. (1999). Developing social competence in children (Choice briefs
Number 3). New York, NY: Columbia University, Institute for Urban and
Minority Education Shelby v. Board of Educ. of LeRoy Cent. School District,
2002 WL 31247602 (N.Y. Sup. Ct., Genesee County, July 9, 2002).
Skiba, R.J., & Peterson, R.L. (2000). School Discipline at a crossroads: From zero
tolerance to early response. Exceptional Children, 66(3), 335-346. Retrieved
on March 26, 2005, from the FirstSearch Database.
Snyder, H.N., & Sickmund, M. (1999). Juvenile offenders and victims: A national
report. Washington, DC: U.S. Department of Justice Programs, Office of
Justice Programs, Office of Juvenile Justice and Delinquency Prevention.
Available online at www.ncjrs.org
Sprague, J., Colvin, G., Irvin, L., & Stieber, S. (1998). Assessing school safety in
Oregon: How do school principals respond? Effective School Practices,
17(2),36-44. Retrieved on January 9, 2005 from
'http://www.nau.edu/ihd/positive/library/sprague4.pdf
State of Hawaii Board of Education. (2003, Oct 23). Minutes o f the Regular
Education, K-12 Committee o f the Whole Meeting. Conducted at the meeting
of the Board of Education, Honolulu, HI.
Stetson, R., Kelly, J., & Stetson, E. (1999). Are schools responsible for a civil
society? (ERIC document reproduction services No. ED 429033)
Stetson, R., Stetson, E. & Kelly, J. (1998). Building a civil society: Are schools
responsible?. Paper presented at the Annual Meeting of the American
Educational Research Association, San Diego, CA.
Titles of the Elementary and Secondary Education Act of 1965, as reauthorized by
the No Child Left Behind Act of 2001. (2001, Jan.). Retrieved August 1,
2004, from
http://www.ed.gov/parents/academic/involve/nclbguide/titles.html
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
130
Thomerson, J. (2000 December). Who’s responsible for school safety?. State
Legislatures Magazine. Retrieved on March 21,2005, from
http://www.ncsl.org/progrmas/pubs/1200scl.htm
U.S. Department of Education. Principal questionnaire 2003-2004 school year:
School survey on crime and safety. National Center for Education Statistics.
OMB# 1850-0761. Available online at
http://nces.ed.gov/programs/crime/principals.asp
Ward, C.M. (1998). Student discipline and alleviating criminal behavior in the inner
city. The Urban Review, 30 (1), 29-48.
Warner, B.S., Weist, M.D., & Krulak, A. (1999). Risk factors for school violence.
Urban Education, 34(1), 52-68.
Webb, L.D., & Norton, M.S. (2003). Human Resource Administration: Personnel
Issues and Needs in Education (4th edition). Columbus, Ohio: Merill Prentice
Hall.
Wilgoren, J. (2005, March 22). Shooting Rampage by Student Leaves 10 Dead on
Reservation. The New York Times. Retrieved on April 10, 2005, from
http://www.nytimes.com
Willert, H.J., & Lenhardt, A.M. (2003). Tackling school violence does take the
whole village. The Educational Forum, 67, 110-118. Retrieved on March 20,
2005 from FirstSearch database.
Wilson, D. (2004). The interface of school climate and school connectedness and
relationships with aggression and victimization. The Journal o f School
Health, 74(7), 293-299.
Zeira, A., Astor, R.A., & Benbenishty, R. (2004). School violence in Israel:
Perceptions of homeroom teachers. School Psychology International, 25(2),
149-166. Retrieved on October 5,2004 from Article First database.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
131
Appendix A: Teacher Questionnaire
Survey of School Violence and Safe School Elements (SSVSSE)
Karin Collier
Ivy Kelling
Leigh Ann Siaosi
University of Southern California
Survey items adapted from sections of the following surveys and resources:
Benbenishty, R., & Astor, R.A. (2005). School violence in context: Culture, neighborhood,
family, school, and gender. New York, NY: Oxford University Press.
Bernard, B. (1991). Fostering resiliency in kids: Protective factors in the family, school, and
community. Portland, OR: Northwest Regional Educational Laboratory. Available
online at http://www.nwrac.org/pub/library/frf_foster.pdf
Eisenberg, M.E., Neumark-Sztainer, D., & Perry, C.L. (2003). Peer harassment, school
connectedness, and academic achievement. Journal of School Health, 73(8), 311-
316. Retrieved on March 12,2005 from EBSCO database.
Furlong, M.J. & Morrision, G. (1998) California School Climate and Safety Survey-
Secondary Version. Available online at http://www.education.ucsb.edu/schpsych/
CSCSS/PDF/CSCSS.v7-12.pdf
Henson, L., Massey, O., & Armstrong, K. (2002). Student perceptions o f school safety:
Results o f the 2001 Omnibus Survey. Tampa, FL: The Louis de la Parte Florida
Mental Health Institute, University of South Florida. FMHI Publication #207-4.
Skiba, R. & Peterson, R. (2003). Teaching the social curriculum: School discipline as
instruction. Preventing School Failure, 47(2), 68-73. Retrieved on March 22, 2005
from EBSCO database.
Stetson, R., Kelly, J., & Stetson, E. (1999). Are schools responsible for a civil society?
(ERIC document reproduction services No. ED 429033)
Wilson, D. (2004). The interface of school climate and school connectedness and
relationships with aggression and victimization. The Journal o f School Health,
74(7), 293-299.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
132
Survey of School Violence and Safe School Elements (SSVSSE)
Teacher Version
Karin Collier, Ivy Kelling, and Leigh Ann Siaosi
Demographics
For the purpose of this study, we need a few details about you and your
school. We remind you again that all our analyses are aggregated so that your
responses are kept confidential.
1.) Gender
I I Female □ Male
2.) Number of years as a teacher:
□ 2 or less
□ 2-3
□ 4-5
□ 6-10
□ 11-20
I I 20 or more
3.) Number of years as a teacher in this school:
□ 2 or less
□ 2-3
□ 4-5
□ 6-10
□ 11-20
I I 20 or more
4.) School level of your current employment:
I I Elementary School
I I Middle/Intermediate School
I I High School
5.) Degree:
I I College/teaching seminar
I I Bachelor’s degree
I I Master’s degree
I I Master’s degree +30 credits
I I Doctorate degree
□ Other
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
133
6.) Type of institution/program where teacher certification was earned:
Institution Type Hawai'i Out of State
State University
Private University
Alternative Certification
Program
—
Not certified
□
Observed Risk Behaviors in School
7.) To what extent do the following behaviors
happen at your school?
Students get into fights.
Students drink alcohol in school.
Students destroy things (vandalism).
Students steal things from other students or teachers.
Students use drugs in school._____________________
Students threaten or bully other students or teachers.
Students bring guns to school.____________________
Students bring other weapons to schools (such as knives,
sticks).
Overall, there is a violent atmosphere in school.
Outsiders (adults) enter the campus during the school
day and threaten, harass, or fight with students and staff.
There is dangerous gang activity in my school._______
Teachers or other staff members come on to, sexually
harass, or bother the students.
Teachers or other staff members hurt students verbally
(insult, humiliate, and curse)._____________________
Teachers or other staff members hurt students physically
(slap, pinch, push).____________________________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Severity o f the School Violence Problem
8.) What is the magnitude of the school violence problem in your school?
I IA very small problem or not a problem at all
I I A small problem
I I A medium-level problem
I IA large problem
I I A very large problem
Violent Behaviors in School
9.) We want to know about violent behaviors among
students that have occurred since the beginning of this
school year. Please consider events that happened in
school, on the way to school, during recess, or during
classes. For each of the following, please estimate the
number incidents that have occurred since the beginning
of this school year.
Students jumped on a student and hurt him or her._______
Students ostracized (socially isolated) a student._________
A student came to school with a gun._________________
A student forcefully took things from another student._____
A student intentionally cut another student with a knife or a
sharp instrument._______________________________
A student kicked or punched another student in order to hurt
him or her.___________________________________
A student mocked, insulted, or humiliated another student
(verbally).____________________________________
A student stole personal belongings or equipment from
another student.________________________________
A student threatened another student with a gun._________
A student threatened another student with a knife.________
A student threatened to hurt or hit another student._______
Gang members threatened and pressured a student._______
There was a conflict between students from different ethnic
groups.______________________________________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
135
Prevention and Intervention
10.) We wish to learn about your activities to
prevent or deal with violence in your class. For each
of the following actions, please indicate how many
times you carried out this activity in your class
during this school year.
Prevention:
You held an educational program in the classroom that dealt I
directly with the subject of violent behavior in school. o
You consulted with the school management on how to
prevent violence in the classroom.
You consulted with the school psychologist on how to
prevent violence in the classroom. o
You consulted with the school counselor on how to prevent
violence in the classroom.
You ran educational activities and class discussions on the
issue. o
You worked with the students’ representatives or a group of |
students from the class on this subject.
You brought up the subject of violence in front of the
parents. O
Responses to a violent incident:
You talked with a student about his or her violent behavior.
You punished a student for his or her violent behavior.
You punished the whole class due to a violent event.
You ran educational activities or discussions in class due to
a violent event.
You sent a student to the principal.
You wrote to a student’s parents or talked to them on the
phone due to the student’s violent behavior. o
You invited a student’s parents to come in for a discussion
of their child’s violent behavior.
You made sure that a student from your class would be
suspended from school due to his or her violent behavior. o
You consulted the counselor.
You consulted the principal and the management team .
You consulted the school psychologist.
You made sure the police got involved.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
School Climate
11.) For each of the following statements, please indicate how well
it describes your school.
This school is badly affected by crime and violence in the community.
Students help make important decisions at this school.
Students are respected at my school.
Preventing violence is a priority at this school.
The teachers tend to ignore “low levels” of violence (e.g., shoves,
threats, and verbal abuse)._______________________________
Dealing with violence is a major priority for the principal.
The principal supports teachers who deal with violence.
The teachers are afraid to break up fights between students.
I have a feeling that I spend too much of my time as a teacher dealing
with problems of violence.______________________________
The school’s leadership acts to reveal and to discuss seriously the
issue of school violence (doesn’t “sweep the problem under the rag”).
I feel helpless when I am dealing with school violence.
The teachers choose to ignore violent acts that are not within their
immediate responsibility.________________________________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
Connectedness
12.) To what extent do the following behaviors happen at
your school?
When students have an emergency, someone is there to help. T V 1 1 V 1 1 k j tU U V U k J U U T V U il V illV l g V U V J j k J V /lllV U llV 1 U U 1 V 1 V v u
Students help make important decisions at this school.
Teachers respect students.
Students respect teachers.
Students here get involved in sports, clubs, and other school
activities.
Teachers let students know when they are doing a good job.
Teachers have close, helpful relationships with students.
At this school, students and teachers really care for each other.
The school makes a genuine effort to involve students in the
prevention of violence.__________________________
Expectations
13.) To what extent do the following behaviors happen at
your school?
At this school, the students are really motivated to learn.
There is an expectation among staff, parents, and students that
students are capable of high achievement._______________
Students are afforded multiple opportunities for success.
Students are given responsibilities related to promoting a safe
school environment.
School-wide expectations are clearly communicated to students.
There is consistency between expectations, rules, and
consequences.
I have spent time in my classroom communicating my
expectations to my students.___________________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Collaborative Community Network
14.) In this part, we ask you to evaluate your relationship with
others who might be relevant to dealing with school violence.
PTA (or other group o f parents):__________________________
Are involved in dealing with violence in my school.
Have a program that involves parents at school helping to promote
school safety._____________________________________
Make a significant contribution to effective coping with violence.
Support staff (counselors, psychologists, SBBH personnel)_____
Have adequate knowledge and skills to help.
Invest the time necessary to address the issue.
Cooperate with me on this issue.
Initiate educational activities and programs in this area.
Make a significant contribution to effective coping with violence.
Resource Teachers and Complex Area Supervisors:___________
Provide the necessary resources for dealing with violence.
Provide knowledge and skills that are needed to deal with
violence._________________________________________
Recognize the importance of the issue.
Understand the safety needs of the school.
Make a significant contribution to effective coping with violence.
DOE State Level Supervisors/Personnel:____________________
Provide the necessary resources for dealing with violence.
Provide knowledge and skills that are needed to deal with
violence._________________________________________
Recognize the importance of the issue.
Understand the safety needs of the school.
Make a significant contribution to effective coping with violence.
Police:________________________________________________
Understand the safety needs of the school.
School involves police in violence prevention efforts.
School involves police in violence intervention efforts.
Make a significant contribution to effective coping with violence.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
139
Discipline and Crisis Plans
15.) To what extent do the following behaviors happen
at your school?
When students break rules, they are treated firmly but
fairly.______________________________________
Teachers are fair.
The rules at this school are fair.
School Leadership apply discipline rules fairly.
There is consistent enforcement of discipline in school.
There are clear school rules regarding discipline and
violence, and these rules are known to teachers.
There are clear school rules regarding discipline and
violence, and these rules are known to students.
16.) Does your school have a written plan that
describes procedures to be performed in the
following crises?
Shootings or armed person on campus
Riots or larger scale fights
Bomb scares, anthrax scares, or comparable school-
wide threats (not including fires)_______________
Natural disasters (e.g. hurricane, tidal wave)
Hostages
Chemical, biological, or radiological threats or
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
Effectiveness o f Discipline Plan Components
17.) How effective have these strategies been in making
your school feel safe?
Suspending/expelling students who commit acts of
violence._________________________________
Putting more security devices in schools.
Having more school resource officers or school safety
managers.________________________________
Training students in conflict resolution and anger
management.______________________________
Training students to resolve conflicts.____________
Training teachers to resolve conflicts.
Training students to accept differences in others.
Keeping drugs out of school.__________________
Having counselors to help students in school.______
Controlling guns in school.___________________
Involving more parents with school.
Leadership training for students.
Teacher Preparation
18.) How prepared do you feel for managing school violence?
I I Not at all
I I A little
C D Some
I I Quite a bit
I I Very much
19.) How prepared do you feel your colleagues (at your school) are to manage
school violence?
I I Not at all
I I A little
I I Some
I I Quite a bit
I I Very much
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
141
20.) Where did you learn strategies for managing school violence?
I I Teacher preparation program
I I Professional development workshops
I I Personal experience (on the job)
I I Reading published materials on the subject (books, articles, journals,
etc.)
I I From other educators
I 1 Other: ____________________
I I Do not possess such knowledge
21.) What is the most appropriate venue for learning strategies to manage
school violence?
1 I Teacher preparation program
I I Professional development workshops
I I Personal experience (on the job)
I I Reading published materials on the subject (books, articles, journals,
etc.)
I I From other educators
I I Other: ____________________
22.) To your best recollection, please indicate the number
of classes during teacher preparation in which the
following topics were addressed.
Discipline and Crisis Plans (including classroom management,
school policies, and emergency procedures)
Expectations (establishing, modeling, and communicating
expectations as a means for reducing problem behaviors)
Collaborative Community Network (strategies for forming
relationships with parents and community partners to ensure a safe
school environment; a team approach to assisting students
exhibiting risk factors of violent behavior)
Connectedness (fostering positive peer relations, enhancing
student-teacher relations, supporting the alienated child)
Positive School Climate (including students in decision making,
assigning students responsiblitiy related to school safety, assessing
and/or improving school climate)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23.) Please indicate how often the following topics were
addresed through professional development activities during
the last school year.
Discipline and Crisis Plans (including classroom management, school
policies, and emergency procedures)
Expectations (establishing, modeling, and communicating expectations
as a means for reducing problem behaviors)_______________________
Collaborative Community Network (strategies for forming
relationships with parents and community partners to ensure a safe
school environment; a team approach to assisting students exhibiting
risk factors of violent behavior)
Connectedness (fostering positive peer relations, enhancing student-
teacher relations, supporting the alienated child)__________________
Positive School Climate (including students in decision making,
assigning students responsiblitiy related to school safety, assessing
and/or improving school climate)
24.) Please indicate whether or not you feel there is a
training need related to each topic listed.
Discipline and Crisis Plans (including classroom management,
school policies, and emergency procedures)_______________
Expectations (establishing, modeling, and communicating
expectations as a means for reducing problem behaviors)_____
Collaborative Community Network (strategies for forming
relationships with parents and community partners to ensure a
safe school environment; a team approach to assisting students
exhibiting risk factors of violent behavior)
Connectedness (fostering positive peer relations, enhancing
student-teacher relations, supporting the alienated child)______
Positive School Climate (including students in decision making,
assigning students responsiblitiy related to school safety,
assessing and/or improving school climate)________________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
143
School Safety Responsibility
25.) How much responsibility for ensuring safe campuses do
you place on:
Administrators?____________________________________
Teachers?________________________________________
Parents?_________________________________________
Students?________________________________________
Community (government/private agencies, law enforcement, etc
Resource Teachers and Complex Area Supervisors?_________
DOE State Level Supervisors/Personnel?_________________
26.) How much responsibility is currently assumed by the
following stakeholders at your school for ensuring safety on
campus?
Administrators_________________________________
Teachers_____________________________________
Parents______________________________________
Students_____________________________________
Community (government/private agencies, law enforcement,
Resource Teachers and Complex Area Supervisors_______
DOE State Level Supervisors/Personnel_______________
27.) In your opinion, who should be held most responsible for ensuring that
schools are safe? Please rank your responses with 1 being most responsible and
7 being least responsible.
Administrators
Teachers
Parents
Students
Community (government/private agencies, law enforcement, etc.)
Resource Teachers and Complex Area Supervisors
DOE State Level Supervisors/Personnel
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Asset Metadata

Creator Kelling, Ivy Kazue (author)

Core Title Elementary teacher perceptions of school violence: Safe school elements and responsibility.

Contributor Digitized by ProQuest (provenance)

School Rossier School of Education

Degree Doctor of Education

Degree Program Education

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Education, administration,education, elementary,OAI-PMH Harvest

Language English

Advisor Sundt, Melora (committee chair), Astor, Ron (committee member), Gothold, Stuart (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-583426

Unique identifier UC11336750

Identifier 3236517.pdf (filename),usctheses-c16-583426 (legacy record id)

Legacy Identifier 3236517.pdf

Dmrecord 583426

Document Type Dissertation

Rights Kelling, Ivy Kazue

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA